-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathindex.html
More file actions
1708 lines (1595 loc) · 165 KB
/
index.html
File metadata and controls
1708 lines (1595 loc) · 165 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<html lang="en">
<head>
<title>Rohit Verma</title>
<meta charset="UTF-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=no">
<meta name="description" content="">
<meta name="keywords" content="">
<meta name="author" content="">
<link rel="shortcut icon" href="images/favicon.ico">
<!-- CSS | STYLE -->
<link rel="stylesheet" type="text/css" href="css/bootstrap.min.css" />
<link rel="stylesheet" type="text/css" href="css/font-awesome.min.css" />
<link rel="stylesheet" type="text/css" href="css/linecons.css" />
<link rel="stylesheet" type="text/css" href="css/normalize.css" />
<link rel="stylesheet" type="text/css" href="css/colors/brown.css" id="csscolors"/>
<link rel="stylesheet" type="text/css" href="css/style.css?v=1.2.0" />
<link rel="stylesheet" type="text/css" href="css/academicons.min.css"/>
<!-- CSS | Google Fonts -->
<link href='https://fonts.googleapis.com/css?family=Montserrat:400' rel='stylesheet' type='text/css'>
<link href='https://fonts.googleapis.com/css?family=Raleway:200,400,300,500,600' rel='stylesheet' type='text/css'>
<noscript>
<style>
@media screen and (max-width: 755px) {
.hs-content-scroller {
overflow: visible;
}
}
</style>
</noscript>
</head>
<body>
<!-- Page preloader -->
<!-- Courtesy: http://www.demo.codiux.com/pulsev2/ -->
<div id="page-loader">
<canvas id="demo-canvas"></canvas>
</div>
<!-- container -->
<div id="hs-container" class="hs-container">
<!-- <div class="switcher" style="left:-40px;">
<a id="switch-panel" class="hide-panel"><i class="li_params"></i></a>
<ul class="colors-list">
<li>
<a title="blue" class="blue"></a>
</li>
<li>
<a title="brown" class="brown"></a>
</li>
<li>
<a title="cyan" class="cyan"></a>
</li>
<li>
<a title="gray" class="gray"></a>
</li>
<li>
<a title="green" class="green"></a>
</li>
<li>
<a title="light-brown" class="light-brown"></a>
</li>
<li>
<a title="light-green" class="light-green"></a>
</li>
<li>
<a title="orange" class="orange"></a>
</li>
</ul>
</div> -->
<!-- <div class="switcher-responsive" style="right:-40px;">
<a id="switch-panel-responsive" class="hide-panel"><i class="li_params"></i></a>
<ul class="colors-list">
<li>
<a title="blue" class="blue"></a>
</li>
<li>
<a title="brown" class="brown"></a>
</li>
<li>
<a title="cyan" class="cyan"></a>
</li>
<li>
<a title="gray" class="gray"></a>
</li>
<li>
<a title="green" class="green"></a>
</li>
<li>
<a title="light-brown" class="light-brown"></a>
</li>
<li>
<a title="light-green" class="light-green"></a>
</li>
<li>
<a title="orange" class="orange"></a>
</li>
</ul>
</div> -->
<!-- Sidebar-->
<div class="aside1">
<!-- <a class="contact-button"><i class="fa fa-paper-plane"></i></a> -->
<!-- <a class="download-button"><i class="fa fa-cloud-download"></i></a> -->
<div class="aside-content"><a href="https://www.intel.in/content/www/in/en/research/overview.html" target="_blank"><span class="part1">Intel</span><span class="part2"></span></a>
</div>
</div>
<aside class="hs-menu" id="hs-menu">
<!-- <canvas id="demo-canvas"></canvas> -->
<!-- Profil Image-->
<div class="hs-headline">
<a id="my-link" href="#my-panel"><i class="fa fa-bars"></i></a>
<!-- <a href="#" class="download"><i class="fa fa-cloud-download"></i></a> -->
<div class="img-wrap">
<img src="images/portrait.jpg" alt="" width="150" height="150" />
</div>
<div class="profile_info">
<h1>Rohit Verma</h1>
<h4>Research Scientist</h4>
<h6>Intel Labs<br>India</h6>
</div>
<div style="clear:both"></div>
</div>
<div class="separator-aside"></div>
<!-- End Profil Image-->
<!-- menu -->
<nav>
<a href="#section1"><span class="menu_name">ABOUT</span><span class="fa fa-home"></span> </a>
<a href="#section2"><span class="menu_name">RESUME</span><span class="fa fa-newspaper-o"></span> </a>
<a href="#section3"><span class="menu_name">PUBLICATIONS</span><span class="fa fa-pencil"></span> </a>
<!-- <a href="#section4"><span class="menu_name">RESEARCH</span><span class="fa fa-laptop"></span> </a> -->
<!-- <a href="#section5"><span class="menu_name">PRICING</span><span class="fa fa-money"></span> </a>
<a href="#section6"><span class="menu_name">SKILLS</span><span class="fa fa-diamond"></span> </a>
<a href="#section7"><span class="menu_name">WORKS</span><span class="fa fa-archive"></span> </a> -->
<!-- <a href="#section8"><span class="menu_name">CONTACT</span><span class="fa fa-paper-plane"></span> </a> -->
</nav>
<!-- end menu-->
<!-- social icons -->
<div class="aside-footer">
<a href="https://www.facebook.com/rohitistheone" target="_blank"><i class="fa fa-facebook"></i></a>
<a href="https://twitter.com/rohitverma246" target="_blank"><i class="fa fa-twitter"></i></a>
<a href="https://in.linkedin.com/in/rohit-verma-00296055" target="_blank"><i class="fa fa-linkedin"></i></a>
<a href="https://github.com/rv355" target="_blank"><i class="fa fa fa-github"></i></a>
<!-- <a href="https://dblp.org/pid/75/1850" target="_blank"><i class="ai ai-dblp"></i></a> -->
<a href="https://scholar.google.com/citations?user=CDTFqc4AAAAJ&hl=en" target="_blank"><i class="ai ai-google-scholar"></i></a>
</div>
<!-- end social icons -->
</aside>
<!-- End sidebar -->
<!-- Go To Top Button -->
<a href="#hs-menu" class="hs-totop-link"><i class="fa fa-chevron-up"></i></a>
<!-- End Go To Top Button -->
<!-- hs-content-scroller -->
<div class="hs-content-scroller">
<!-- Header -->
<div id="header_container">
<div id="header">
<div><a class="home"><i class="fa fa-home"></i></a>
</div>
<div><a href="" class="previous-page arrow"><i class="fa fa-angle-left"></i></a>
</div>
<div><a href="" class="next-page arrow"><i class="fa fa-angle-right"></i></a>
</div>
<!-- News scroll -->
<div class="news-scroll">
<span><i class="fa fa-line-chart"></i>RECENT ACTIVITY : </span>
<ul id="marquee" class="marquee">
<li>
Dec 2025: 'Direct File Path Access with NVMe over Fabrics for High-Performance Remote Data Reads' accepted at INFOCOM 2026
</li>
<li>
Jun 2025: 'Towards An Improved Video RAG Workflow With Orchestration Support in A Visual Data Management System' accepted at DEEM @ SIGMOD 2025
</li>
<li>
Jan 2025: 'Unveiling the Trade-offs: A Parameter-Centric Comparison of Synchronous and Asynchronous Federated Learning' accepted at COMSNETS 2025
</li>
<li>
Aug 2024: 'Selective Graph Convolutional Network for Efficient Routing' accepted at AIMLSystem 2024.
</li>
<li>
Aug 2024: 'On-the-Go Automated Break Recommendation for Stress Avoidance during Highway Driving' accepted at ACM Journal on Autonomous Transportation Systems.
</li>
</ul>
</div>
<!-- End News scroll -->
</div>
</div>
<!-- End Header -->
<!-- hs-content-wrapper -->
<div class="hs-content-wrapper">
<!-- About section -->
<article class="hs-content about-section" id="section1">
<span class="sec-icon fa fa-home"></span>
<div class="hs-inner">
<!-- <span class="before-title">.01</span> -->
<h2>ABOUT</h2>
<span class="content-title">PERSONAL DETAILS</span>
<div class="aboutInfo-contanier">
<div class="about-card">
<!-- <div class="face2 card-face">
<div id="cd-google-map">
<div id="google-container"></div>
<div id="cd-zoom-in"></div>
<div id="cd-zoom-out"></div>
<address>Computer Laboratory, University of Cambridge, UK</address>
<div class="back-cover" data-card-back="data-card-back"><i class="fa fa-long-arrow-left"></i>
</div>
</div>
</div> -->
<div class="face1 card-face">
<div class="about-cover card-face">
<!-- <a class="map-location" data-card-front="data-card-front"><img src="images/map-icon.png" alt="">
</a> -->
<div class="about-details">
<!-- <div><span class="fa fa-inbox"></span><span class="detail">rohit [dot] verma [at] cse [dot] iitkgp [dot] ac [dot] in</span>
</div> -->
<div><span class="fa fa-inbox"></span><span class="detail">rohit1.verma@intel.com</span>
</div>
<div><span class="fa fa-paper-plane"></span><span class="detail">SRR2, Intel Technology India Pvt. Ltd<br> Bangalore, India</span>
</div>
</div>
<div class="cover-content-wrapper">
<span class="about-description">Hello. I am a<span class="rw-words">
<span><strong>Programmer</strong></span>
<span><strong>Researcher</strong></span>
<span><strong>Programmer</strong></span>
<span><strong>Researcher</strong></span>
<span><strong>Programmer</strong></span>
</span>
<br>I am passionate about developing systems.
<!-- <br>Welcome to my Personal profile</span> -->
<!-- <span class="status">
<span class="fa fa-circle"></span> -->
<!-- <span class="text">Available as <strong>freelance</strong></span> -->
</span>
</div>
</div>
</div>
</div>
<div class="more-details">
<div class="tabbable tabs-vertical tabs-left">
<ul id="myTab" class="nav nav-tabs">
<li class="active">
<a href="#bio" data-toggle="tab"><b>Bio</b></a>
</li>
<li>
<a href="#hobbies" data-toggle="tab"><b>Research</b></a>
</li>
<!-- <li>
<a href="#facts" data-toggle="tab">Facts</a>
</li> -->
</ul>
<div id="myTabContent" class="tab-content">
<div class="tab-pane fade in active" id="bio">
<p>I am a Systems Research Scientist in the Edge Computing Group at <a href="https://www.intel.in/content/www/in/en/research/overview.html">Intel</a>, India. Prior to joining Intel, I was working as a Research Associate in the <a href="https://www.cl.cam.ac.uk/research/srg/">Systems Research Group</a> at the <a href="https://www.cst.cam.ac.uk/">Department of Computer Science and Technology</a>, <a href="https://www.cam.ac.uk/">University of Cambridge</a><br>
I completed my PhD from the <a href="http://www.cnergres.iitkgp.ac.in/">Complex Network Research Group</a> at the <a href="http://cse.iitkgp.ac.in/">Department of Computer Science and Engineering</a>, <a href="http://iitkgp.ac.in/">Indian Institute of Technology, Kharagpur</a> under the supervision of <a href="http://cse.iitkgp.ac.in/~sandipc/">Dr. Sandip Chakraborty</a> and <a href="https://cse.iitkgp.ac.in/~bivasm/">Dr. Bivas Mitra</a> (2016-2020). I am a recipient of the <strong>COMSNETS Association Best PhD Thesis Award 2021</strong>.</p>
</div>
<div class="tab-pane fade" id="hobbies">
<p>My primary area of research has been in the field of sensor data collection and analysis obtained from multi-modal sources. In my PhD I had been utilizing such information towards developing systems supporting the transport systems. Currently, I am more focused on the real-time aspect of the data being generated by sensors deployed at a citywide scale. The idea being not only to store and learn from the data that these citywide sensors generate but to take crucial decisions with minimum latency to support the needs of the city.</p>
<div style="clear:both;"></div>
</div>
</div>
</div>
</div>
</div>
</div>
<br>
</article>
<!-- End About Section -->
<!-- Resume Section -->
<article class="hs-content resume-section" id="section2">
<span class="sec-icon fa fa-newspaper-o"></span>
<div class="hs-inner">
<!-- <span class="before-title">.02</span> -->
<h2>RESUME
<!-- <a href="pdfs/resume.pdf" target="_blank"><i class="fa fa-cloud-download"></i></a> -->
</h2>
<div>
<a href="#experience" class="buttonresume">EXPERIENCE</a>
<a href="#commitees" class="buttonresume">COMMITTEES</a>
<a href="#talks" class="buttonresume">TALKS</a>
<a href="#teaching" class="buttonresume">TEACHING</a>
<a href="#recognition" class="buttonresume">RECOGNITIONS</a>
</div>
<!-- <br><br> -->
<!-- Resume Wrapper -->
<div class="resume-wrapper">
<ul class="resume">
<!-- Resume timeline -->
<li class="time-label">
<span class="content-title">CURRENT</span>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Bangalore, India</span>
<h3 class="timeline-header">SYSTEMS RESEARCH SCIENTIST</h3>
<div class="timeline-body">
<h4>Intel, India</h4>
</div>
</div>
</li>
<li class="time-label">
<span class="content-title">EDUCATION</span>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Kharagpur, India</span>
<h3 class="timeline-header">Doctor of Philosophy (PhD) - Computer Science and Engineering</h3>
<div class="timeline-body">
<h4>INDIAN INSTITUTE OF TECHNOLOGY (2016 - 2020)</h4>
<span><strong>Thesis Title:</strong> Spatio-Temporal Data Collection and Analysis for Developing Transport related Services<br><strong>Supervisor:</strong> <a href="http://cse.iitkgp.ac.in/~sandipc/" target="_blank">Dr. Sandip Chakraborty</a>, <a href="http://cse.iitkgp.ac.in/~bivasm/" target="_blank">Dr. Bivas Mitra</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Durgapur, India</span>
<h3 class="timeline-header">Bachelor of Technology (B.Tech) - Computer Science and Engineering</h3>
<div class="timeline-body">
<h4>NATIONAL INSTITUTE OF TECHNOLOGY (2009 - 2013)</h4>
</div>
</div>
</li>
<li class="time-label" id="experience">
<span class="content-title">PROFESSIONAL EXPERIENCE</span>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Surat, India</span>
<h3 class="timeline-header">ADJUNCT FACULTY</h3>
<div class="timeline-body">
<h4>INDIAN INSTITUTE OF INFORMATION TECHNOLOGY (2024-)</h4>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Bangalore, India</span>
<h3 class="timeline-header">EXECUTIVE COMMITTEE MEMBER</h3>
<div class="timeline-body">
<h4>IEEE Computer Society Bangalore Chapter (2023-)</h4>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Cambridge, UK</span>
<h3 class="timeline-header">RESEARCH ASSOCIATE</h3>
<div class="timeline-body">
<h4>UNIVERSITY OF CAMBRIDGE (2020-2022)</h4>
<span>Project: <a href="http://pages.cdbb.uk/" target="_blank">CDBB Digital Twin Program</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Kharagpur, India</span>
<h3 class="timeline-header">JUNIOR RESEARCH FELLOW</h3>
<div class="timeline-body">
<h4>INDIAN INSTITUTE OF TECHNOLOGY (2015)</h4>
<span>Project: <a href="https://www.meity.gov.in/content/projects-11" target="_blank">DiSARM</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Bangalore, India</span>
<h3 class="timeline-header">SOFTWARE DEVELOPER</h3>
<div class="timeline-body">
<h4>SCHNEIDER ELECTRIC INDIA PVT. LTD. (2013 - 2015)</h4>
<span>.NET and ANDROID Development</span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Kharagpur, India</span>
<h3 class="timeline-header">INTERN</h3>
<div class="timeline-body">
<h4>INDIAN INSTITUTE OF TECHNOLOGY (2012)</h4>
<span><b>Guide:</b> <a href="http://www.facweb.iitkgp.ernet.in/~niloy/" target="_blank">Dr. Niloy Ganguly</a><br><b>Project Title:</b> Analyzing a Fastest Path Algorithm for Spatio-Temporal Networks</span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Kharagpur, India</span>
<h3 class="timeline-header">INTERN</h3>
<div class="timeline-body">
<h4>INDIAN INSTITUTE OF TECHNOLOGY (2011)</h4>
<span><b>Guide:</b> <a href="http://www.facweb.iitkgp.ernet.in/~niloy/" target="_blank">Dr. Niloy Ganguly</a><br><b>Project Title:</b> Delay Tolerant Network and its Application in Post Disaster Management</span>
</div>
</div>
</li>
<li class="time-label" id="commitees"></li>
<span class="content-title">CONFERENCE COMMITEE MEMBER</span>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Bangalore, India</span>
<h3 class="timeline-header">POSTER CO-CHAIR</h3>
<div class="timeline-body">
<h4>COMSNETS 2022-2026</h4>
<span>WebPage: <a href="http://www.comsnets.org/" target="_blank">COMSNETS</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Bangalore, India</span>
<h3 class="timeline-header">SysAI Workshop CO-CHAIR</h3>
<div class="timeline-body">
<h4>SysAI Workshop, COMSNETS 2025, 2026</h4>
<span>WebPage: <a href="http://www.comsnets.org/" target="_blank">SysAI Workshop, COMSNETS</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Bangalore, India</span>
<h3 class="timeline-header">TPC CO-CHAIR</h3>
<div class="timeline-body">
<h4>ICDDS 2025</h4>
<span>WebPage: <a href="https://icdds.org/" target="_blank">ICDDS</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Hyderabad, India</span>
<h3 class="timeline-header">FINANCE CO-CHAIR</h3>
<div class="timeline-body">
<h4>ICDCN 2025</h4>
<span>WebPage: <a href="https://cse.iith.ac.in/icdcn-2025" target="_blank">ICDCN 2025</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Bangalore, India</span>
<h3 class="timeline-header">TRACK CO-CHAIR</h3>
<div class="timeline-body">
<h4>IEEE CONECCT 2024, 2025 - Computer Engineering & AI Track</h4>
<span>WebPage: <a href="https://ieee-conecct.org/" target="_blank">IEEE CONECCT</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Bangalore, India</span>
<h3 class="timeline-header">PUBLICITY CO-CHAIR</h3>
<div class="timeline-body">
<h4>COMSNETS 2021</h4>
<span>WebPage: <a href="http://www.comsnets.org/" target="_blank">COMSNETS 2021</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Kharagpur, India</span>
<h3 class="timeline-header">WEBMASTER</h3>
<div class="timeline-body">
<h4>IMOBILE: The India Chapter of ACM SIGMOBILE</h4>
<span>WebPage: <a href="http://imobile.acm.org/" target="_blank">IMOBILE</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>New Delhi, India</span>
<h3 class="timeline-header">WEB CHAIR</h3>
<div class="timeline-body">
<h4>MobiCom 2018</h4>
<span>WebPage: <a href="https://sigmobile.org/mobicom/2018/" target="_blank">MobiCom 2018</a></span>
</div>
</div>
</li>
<li>
<div class="timeline-item">
<span class="timeline-location"><i class="fa fa-map-marker"></i>Bangalore, India</span>
<h3 class="timeline-header">WEB CHAIR</h3>
<div class="timeline-body">
<h4>COMSNETS 2018</h4>
<span>WebPage: <a href="http://www.comsnets.org/" target="_blank">COMSNETS</a></span>
</div>
</div>
</li>
<li class="time-label" id="talks"></li>
<span class="content-title">TALKS</span>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header">Invited talk at the <b>IEEE International Conference on Contemporary Computing and Communications (InC4) (Mar 2025).</b></h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header">Invited talk at the <b>Research Scholar's Day RSD 2022 at National Institute of Technology Durgapur (Nov 2022).</b></h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header">Invited talk at the <b>Workshop on "Last-mile" Challenges and Standardization Opportunities in Smart Infrastructure, COMSNETS 2022 (Jan 2022).</b></h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header">Technical talk session hosted by <b>IEEE Communication Society- Delhi Chapter and IEEE Computer Society- Delhi Chapter (Aug 2021).</b></h3>
</div>
</li>
<li class="time-label" id="teaching"></li>
<span class="content-title">TEACHING</span>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>Supervisor - Operating Systems</b>, University of Cambridge (Lent 2022)</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>Supervisor - Computer Networking</b>, University of Cambridge (Lent 2022)</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>Supervisor - Computer Networking</b>, University of Cambridge (Lent 2021)</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>TA - Computing Lab</b>, Dept of CSE, IIT Kharagpur (Spring 2018 and 2019)</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>TA - Smartphone Computing</b>, Dept of CSE, IIT Kharagpur (Autumn 2017)</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>TA - Computer Networks Lab</b>, Dept of CSE, IIT Kharagpur (Autumn 2016)</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>TA - Programming and Data Structure</b>, Dept of CSE, IIT Kharagpur (Spring 2016, Spring 2017, Autumn 2018)</h3>
</div>
</li>
<li class="time-label" id="recognition"></li>
<span class="content-title">RECOGNITIONS</span>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>COMSNETS Association Best Thesis Award</b> - 2021</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>TCS PhD Fellowship</b> - 2016</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>Google Student Travel Grant</b> for IEEE PerCom 2019</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>LRN Travel Grant</b> for IEEE PerCom 2019</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>Google Student Travel Grant</b> for ACM SIGSPATIAL 2018</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>XRCI Travel Grant</b> for IEEE INFOCOM 2016</h3>
</div>
</li>
<li>
<div class="timeline-item">
<h3 class="timeline-header"><b>ACM IARCS Travel Grant</b> for IEEE INFOCOM 2016</h3>
</div>
</li>
<!-- End Resume timeline -->
</ul>
</div>
<!-- End Resume Wrapper -->
</div>
</article>
<!-- End Resume Section-->
<article class="hs-content publication-section" id="section3">
<span class="sec-icon fa fa-pencil"></span>
<div class="hs-inner">
<!-- <span class="before-title">.03</span> -->
<h2>PUBLICATIONS</h2>
<span class="content-title">PUBLICATIONS LIST</span>
<!-- <div class="row publication-form">
<div class="col-md-6 publication-filter">
<div class="card-drop">
<a class='toggle'>
<i class='icon-suitcase'></i>
<span class='label-active'>ALL</span>
</a>
<ul id="filter">
<li class='active'><a data-label="ALL" data-group="all">ALL</a>
</li>
<li><a data-label="JOURNAL PAPERS" data-group="JOURNAL PAPERS">JOURNAL PAPERS</a>
</li>
<li><a data-label="CONFERENCES" data-group="CONFERENCES">CONFERENCES</a>
</li>
<li><a data-label="DEMONSTRATIONS" data-group="DEMONSTRATIONS">DEMONSTRATIONS</a>
</li>
<li><a data-label="THESES" data-group="THESES">THESES</a>
</li>
<li><a data-label="BOOK CHAPTERS" data-group="BOOK CHAPTERS">BOOK CHAPTERS</a>
</li>
<li><a data-label="BOOK" data-group="BOOK">BOOK</a>
</li>
</ul>
</div>
</div>
<div class="col-md-6 publication-sort">
<div class="sorting-button">
<span>SORTING BY DATE</span>
<button class="desc"><i class="fa fa-sort-numeric-desc"></i>
</button>
<button class="asc"><i class="fa fa-sort-numeric-asc"></i>
</button>
</div>
</div>
</div> -->
<!-- End Filter/Sort Menu -->
<!-- publication wrapper -->
<div id="mygrid">
<!-- publication item -->
<div class="publication_item" data-groups='["all","CONFERENCES"]' data-date-publication="2025-12-08">
<div class="media">
<a href=".publication-detail32" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>Direct File Path Access with NVMe over Fabrics for High-Performance Remote Data Reads</h3>
<h4>Tokyo, Japan</h4>
<span class="publication_description">IEEE INFOCOM 2026</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Conferences</span>
<!-- <span class="label pdf"><a href="https://ieeexplore.ieee.org/document/10885640" target="_blank"><font weight='bold' color="white">PDF</font></a></span> -->
<span class="publication_authors"><b>Rohit Verma</b>, Arun Raghunath</span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail32 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">Direct File Path Access with NVMe over Fabrics for High-Performance Remote Data Reads</h3>
<span class="publication_authors"><b>Rohit Verma</b>, Arun Raghunath</span>
<span class="label label-success">Conferences</span>
<p class="project_desc">With AI workloads highly relying on massive distributed datasets, the performance of remote file access has become crucial. Traditional remote file access mechanisms introduce significant overhead when retrieving data from storage, limiting performance in high-throughput applications. NVMe over Fabrics (NVMe-oF) is a remote file access mechanism that is designed for high-speed, low-latency access to remote block storage, making it a superior alternative to traditional solutions. However, NVMe is inherently a block-based protocol, meaning it does not natively support file names, directory structures, or POSIX semantics. In this paper, we design an approach to enable direct file-path reads over NVMe-oF, bridging the gap between high-performance block devices and file-oriented applications. Our approach streamlines data retrieval by leveraging NVMe's high-speed architecture and low-latency capabilities, reducing unnecessary kernel interactions and optimizing I/O operations. Benchmark results show up to 2-14X improvement in remote read latency compared to existing approaches.</p>
</div>
<!-- <a class="ext_link" href="https://dl.acm.org/doi/abs/10.1145/3735654.3735945" target="_blank"><i class="fa fa-external-link"></i></a> -->
<div style="clear:both"></div>
</div>
<br>
<div class="publication_item" data-groups='["all","CONFERENCES"]' data-date-publication="2025-06-22">
<div class="media">
<a href=".publication-detail31" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>Towards An Improved Video RAG Workflow With Orchestration Support in A Visual Data Management System</h3>
<h4>Berlin, Germany</h4>
<span class="publication_description">Proceedings of the Workshop on Data Management for End-to-End Machine Learning (DEEM 2025 @ SIGMOD)</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Conferences</span>
<!-- <span class="label pdf"><a href="https://ieeexplore.ieee.org/document/10885640" target="_blank"><font weight='bold' color="white">PDF</font></a></span> -->
<span class="publication_authors">Sourish Chatterjee, <b>Rohit Verma</b>, Abhinav Kumar, Arun Raghunath</span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail31 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">Towards An Improved Video RAG Workflow With Orchestration Support in A Visual Data Management System</h3>
<span class="publication_authors">Sourish Chatterjee, <b>Rohit Verma</b>, Abhinav Kumar, Arun Raghunath</span>
<span class="label label-success">Conferences</span>
<p class="project_desc">Video Retrieval-Augmented Generation (RAG) Workflows augment user query contexts with stored video contexts to guide a Large Language Model (LLM) in providing more context-relevant answers. The challenge is to design a workflow where the video contexts can be generated with practical latency values. It is also important to ensure that the resources available to run the workflows are not under-utilized whenever possible. In this paper, we propose a video RAG workflow that utilizes the Visual Data Management System (VDMS) interfaced with the Kubernetes (K8s) orchestration framework to design an enhanced video RAG workflow (VRAG) for faster video context generation in the pre-processing phase and a faster response generation with no impact on response accuracy. Our experiments showcase that compared to Conventional RAG(CRAG) workflows, VRAG reduces the pre-processing latency by a minimum of 10%, which decreases manifold with parallelization and the response latency by a minimum of 70% to as high as 97% with further parallelization.</p>
</div>
<a class="ext_link" href="https://dl.acm.org/doi/abs/10.1145/3735654.3735945" target="_blank"><i class="fa fa-external-link"></i></a>
<div style="clear:both"></div>
</div>
<br>
<div class="publication_item" data-groups='["all","CONFERENCES"]' data-date-publication="2024-12-05">
<div class="media">
<a href=".publication-detail30" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>Unveiling the Trade-offs: A Parameter-Centric Comparison of Synchronous and Asynchronous Federated Learning</h3>
<h4>Bangalore, India</h4>
<span class="publication_description">Proceedings of the 17th International Conference on COMmunication Systems and NETworkS (COMSNETS 2025)</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Conferences</span>
<span class="label pdf"><a href="https://ieeexplore.ieee.org/document/10885640" target="_blank"><font weight='bold' color="white">PDF</font></a></span>
<span class="publication_authors">Satvik Rajesh, Ishaan Gakhar, Rajeev Shorey, <b>Rohit Verma</b></span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail30 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">Unveiling the Trade-offs: A Parameter-Centric Comparison of Synchronous and Asynchronous Federated Learning</h3>
<span class="publication_authors">Satvik Rajesh, Ishaan Gakhar, Rajeev Shorey, <b>Rohit Verma</b></span>
<span class="label label-success">Conferences</span>
<p class="project_desc">Federated Learning (FL) has emerged as a transformative paradigm enabling decentralized model training while preserving data privacy. Within FL, synchronous and asynchronous training modes are two widely adopted approaches. This paper compares synchronous and asynchronous federated learning by examining their impact across accuracy and convergence epochs. Through a series of experimental evaluations, we demonstrate the strengths and limitations of each approach in diverse settings, offering insights into the trade-offs and applicability of synchronous and asynchronous FL modes for various distributed learning scenarios. Our findings aim to guide researchers and practitioners in selecting optimal FL configurations based on their system requirements and performance priorities.</p>
</div>
<a class="ext_link" href="https://ieeexplore.ieee.org/document/10885640" target="_blank"><i class="fa fa-external-link"></i></a>
<div style="clear:both"></div>
</div>
<br>
<div class="publication_item" data-groups='["all","CONFERENCES"]' data-date-publication="2024-08-14">
<div class="media">
<a href=".publication-detail29" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>Selective Graph Convolutional Network for Efficient Routing</h3>
<h4></h4>
<span class="publication_description">Fourth International Conference on AI-MLSystems</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Conferences</span>
<span class="label pdf"><a href="https://dl.acm.org/doi/10.1145/3703412.3704451" target="_blank"><font weight='bold' color="white">PDF</font></a></span>
<span class="publication_authors">K Nikita, Hayagreev J, <strong>Rohit Verma</strong>, Rajeev Shorey </span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail29 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">Selective Graph Convolutional Network for Efficient Routing</h3>
<span class="publication_authors">K Nikita, Hayagreev J, <strong>Rohit Verma</strong>, Rajeev Shorey</span>
<span class="label label-success">Conferences</span>
<p class="project_desc">Modern-day networks require highly intelligent and dynamic routing policies that are capable of routing data efficiently over a wide range of changing topology and load conditions. Machine learning frameworks like graph neural networks and multi-agent reinforcement learning have been applied to routing scenarios that deal with heterogeneous graphs. However, these solutions cannot sustain optimality under higher packet loads because deep Q networks and graph neural networks for a single agent indiscriminately factor in all neighbouring agents during the training phase. This also leads to higher computational costs during training and implementation.
<br>
In this paper, we propose a Selective Graph Convolutional Network (Sel_GCN ) model using Multi-Agent Deep Q Networks and Attention to maximize throughput and minimize latency over various packet load conditions. We evaluate our technique on a simulator and compare it to previous implementations of reinforcement learning and graph neural networks. Our results show that Sel_GCN achieves 100% packet delivery while reducing the CPU and memory usage by about 50% and 45% respectively.</p>
</div>
<a class="ext_link" href="https://dl.acm.org/doi/10.1145/3703412.3704451" target="_blank"><i class="fa fa-external-link"></i></a>
<div style="clear:both"></div>
</div>
<br>
<div class="publication_item" data-groups='["all","JOURNAL PAPERS"]' data-date-publication="2024-08-14">
<div class="media">
<a href=".publication-detail28" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>On-the-Go Automated Break Recommendation for Stress Avoidance<br> during Highway Driving</h3>
<h4></h4>
<span class="publication_description">ACM Journal on Autonomous Transportation Systems</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Journal Paper</span>
<span class="label pdf"><a href="https://dl.acm.org/doi/10.1145/3689438" target="_blank"><font weight='bold' color="white">PDF</font></a></span>
<span class="publication_authors"><strong>Rohit Verma</strong>, Bivas Mitra, Sandip Chakraborty</span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail28 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">On-the-Go Automated Break Recommendation for Stress Avoidance during Highway Driving</h3>
<span class="publication_authors"><strong>Rohit Verma</strong>, Bivas Mitra, Sandip Chakraborty</span>
<span class="label label-success">Journal Paper</span>
<p class="project_desc">Continuous cab driving is considered one of the highly stressful jobs, although the drivers ignore that many a time. Taking a break
from manual driving or transferring the control to another driver to release the stress would be an easy, intuitive solution, although
the challenge is to detect the driving stress while the trip is going on. As driving stress depends on multiple diverse environmental and
affective features, we, in this paper, develop a novel assistive system, SmartHalt, which continuously senses the driving environment
(like road type, congestion, driver's driving pattern, etc.) and then utilizes a spatial time series model of the driving environment with
a deep learning framework to predict whether the driver will be stressed while on the trip. The model also considers the personality
traits of the drivers along with the spatio-temporal features to differentiate the impact of stress on the driving behaviour for different
drivers and recommends taking a break soon before the driving behaviour drops below a critical level. A thorough analysis of the
model over 7 different drivers for a 10 month-long experiment over 204871 km of driving data reveals that the proposed approach
can significantly improve driving behaviour by recommending a driving break at proper times. Following the recommendation by
SmartHalt improves the driving score by ≈ 50% and reduces the number of driving offences by ≈ 50%. SmartHalt can help develop
advanced driving assisting system (ADAS) platforms that understand the affective states of the driver and thus can be helpful for
semi-autonomous driving environments for effective driver-vehicle interactions.</p>
</div>
<a class="ext_link" href="https://dl.acm.org/doi/10.1145/3689438" target="_blank"><i class="fa fa-external-link"></i></a>
<div style="clear:both"></div>
</div>
<br>
<div class="publication_item" data-groups='["all","PREPRINT"]' data-date-publication="2024-03-08">
<div class="media">
<a href=".publication-detail27" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>Towards a Flexible Scale-out Framework for Efficient Visual Data Query Processing</h3>
<h4></h4>
<span class="publication_description">arXiv</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Preprint</span>
<span class="label pdf"><a href="https://arxiv.org/abs/2402.03283" target="_blank"><font weight='bold' color="white">PDF</font></a></span>
<span class="publication_authors"><strong>Rohit Verma</strong>, Arun Raghunath</span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail27 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">Towards a Flexible Scale-out Framework for Efficient Visual Data Query Processing</h3>
<span class="publication_authors"><strong>Rohit Verma</strong>, Arun Raghunath</span>
<span class="label label-success">Preprint Paper</span>
<p class="project_desc">There is growing interest in visual data management systems that support queries with specialized operations ranging from resizing an image to running complex machine learning models. With a plethora of such operations, the basic need to receive query responses in minimal time takes a hit, especially when the client desires to run multiple such operations in a single query. Existing systems provide an ad-hoc approach where different solutions are clubbed together to provide an end-to-end visual data management system. Unlike such solutions, the Visual Data Management System (VDMS) natively executes queries with multiple operations, thus providing an end-to-end solution. However, a fixed subset of native operations and a synchronous threading architecture limit its generality and scalability.
<br>
In this paper, we develop VDMS-Async that adds the capability to run user-defined operations with VDMS and execute operations within a query on a remote server. VDMS-Async utilizes an event-driven architecture to create an efficient pipeline for executing operations within a query. Our experiments have shown that VDMS-Async reduces the query execution time by 2-3X compared to existing state-of-the-art systems. Further, remote operations coupled with an event-driven architecture enables VDMS-Async to scale query execution time linearly with the addition of every new remote server. We demonstrate a 64X reduction in query execution time when adding 64 remote servers.</p>
</div>
<a class="ext_link" href="https://arxiv.org/abs/2402.03283" target="_blank"><i class="fa fa-external-link"></i></a>
<div style="clear:both"></div>
</div>
<br>
<div class="publication_item" data-groups='["all","CONFERENCES"]' data-date-publication="2023-12-05">
<div class="media">
<a href=".publication-detail26" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>Towards an Efficient Federated Learning Framework with Selective Aggregation</h3>
<h4>Bangalore, India</h4>
<span class="publication_description">Proceedings of the 16th International Conference on COMmunication Systems and NETworkS (COMSNETS 2024)</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Conferences</span>
<span class="label pdf"><a href="https://ieeexplore.ieee.org/document/10426966/" target="_blank"><font weight='bold' color="white">PDF</font></a></span>
<span class="publication_authors">Anirudha Kulkarni, Abhinav Kumar, Rajeev Shorey, <b>Rohit Verma</b></span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail26 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">Towards an Efficient Federated Learning Framework with Selective Aggregation</h3>
<span class="publication_authors">Anirudha Kulkarni, Abhinav Kumar, Rajeev Shorey, <b>Rohit Verma</b></span>
<span class="label label-success">Conferences</span>
<p class="project_desc">Federated Learning shows promise for collaborative, decentralized machine learning but faces efficiency challenges; primarily network straggler-induced latency bottlenecks and the need for complex aggregation techniques. To address these issues, ongoing research explores asynchronous FL, i.e., federated learning models, including an Asynchronous Parallel Federated Learning framework. This study investigates the impact of varying worker node numbers on key metrics. More nodes offer faster convergence but may increase communication overhead and straggler vulnerability. We aim to quantify how the number of worker node variations for one global aggregation can affect convergence speed, communication efficiency, model accuracy, and system robustness, optimizing asynchronous FL system configurations. This work is crucial for practical and scalable FL applications, mitigating network stragglers, data distribution, and security challenges. This work introduces Asynchronous Parallel Federated Learning, showcasing a paradigm shift by selectively aggregating early arriving worker node updates with a novel parameter `x', improving efficiency and reshaping FL applications.</p>
</div>
<a class="ext_link" href="https://ieeexplore.ieee.org/document/10426966/" target="_blank"><i class="fa fa-external-link"></i></a>
<div style="clear:both"></div>
</div>
<br>
<div class="publication_item" data-groups='["all","CONFERENCES"]' data-date-publication="2022-11-01">
<div class="media">
<a href=".publication-detail25" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>FedNSE: Optimal Node Selection for Federated Learning with<br> Non-IID Data</h3>
<h4>Bangalore, India</h4>
<span class="publication_description">Proceedings of the 15th International Conference on COMmunication Systems and NETworkS (COMSNETS 2023)</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Conferences</span>
<span class="label pdf"><a href="https://ieeexplore.ieee.org/document/10041396" target="_blank"><font weight='bold' color="white">PDF</font></a></span>
<span class="publication_authors">Sourav Bansal, Manav Bansal, <b>Rohit Verma</b>, Rajeev Shorey and Huzur Saran</span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail25 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">FedNSE: Optimal Node Selection for Federated Learning with Non-IID Data</h3>
<span class="publication_authors">Sourav Bansal, Manav Bansal, <b>Rohit Verma</b>, Rajeev Shorey and Huzur Saran</span>
<span class="label label-success">Conferences</span>
<p class="project_desc">Federated Learning relies heavily on the data available at the worker nodes. In a majority of practical use-cases, the data at the worker nodes are Non-IID (non-independent-and-identically-distributed) in nature. The central server often only selects a subset of the worker nodes to compute the global model. If the subset has more nodes with heterogeneous data, the overall performance would be compromised. Selecting an optimal subset of worker nodes from the available worker nodes is crucial for improving the accuracy and convergence time of Federated Learning models. In this paper, we analyse the impact of Non-IID data, measured using quantity skewness, on the performance of the federated learning model and design approaches to detect the level of skewness of the data distribution at the worker nodes. Following this, we design an approach Federated Node Selection with Entropy (FedNSE) to select the optimal set of worker nodes to improve the accuracy and convergence time of a federated learning model. Our experiments performed in a simulated Federated Learning environment show that using FedNSE that selects the optimal subset of worker nodes improves the convergence time and accuracy compared to the existing approaches.</p>
</div>
<a class="ext_link" href="https://ieeexplore.ieee.org/document/10041396" target="_blank"><i class="fa fa-external-link"></i></a>
<div style="clear:both"></div>
</div>
<br>
<div class="publication_item" data-groups='["all","CONFERENCES"]' data-date-publication="2022-11-01">
<div class="media">
<a href=".publication-detail24" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>Multi-Agent Packet Routing (MAPR): Co-Operative Packet Routing Algorithm with Multi-Agent Reinforcement Learning</h3>
<h4>Bangalore, India</h4>
<span class="publication_description">Proceedings of the 15th International Conference on COMmunication Systems and NETworkS (COMSNETS 2023)</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Conferences</span>
<span class="label pdf"><a href="https://ieeexplore.ieee.org/document/10041337" target="_blank"><font weight='bold' color="white">PDF</font></a></span>
<span class="publication_authors">Aniket Modi, Rishi Shah, Krishnanshu Jain, <b>Rohit Verma</b>, Rajeev Shorey and Huzur Saran</span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail24 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">Multi-Agent Packet Routing (MAPR): Co-Operative Packet Routing Algorithm with Multi-Agent Reinforcement Learning</h3>
<span class="publication_authors">Aniket Modi, Rishi Shah, Krishnanshu Jain, <b>Rohit Verma</b>, Rajeev Shorey and Huzur Saran</span>
<span class="label label-success">Conferences</span>
<p class="project_desc">Packet Routing has been investigated intensively in computer networking with numerous existing solutions. However, with the increasing usage of distributed networks that have frequent changes in the network architecture and the environment, there is a need of solutions that learn changes in the network and modify the routing policy in real time. Multi-Agent Reinforcement Learning (MARL) is emerging as a promising approach for such distributed networks. But, existing MARL based approaches are limited in the level of co-operation amongst agents and use Deep Q-Networks (DQNs) that could easily lead to overestimation bias. Moreover, for a co-operative learning scenario such as MARL, it is crucial for the reward to be generated in a way that agents that could hamper the network performance are not created.
<br>
In this paper, we design a packet routing algorithm Multi-Agent Packet Routing (MAPR) that ensures that the neighbouring agent information is captured temporally throughout the training phase. MAPR uses a Double Deep Q Network (DDQN) to ensure that the overestimation bias is avoided in the learning phase. Furthermore, we design a delay-tolerant global-cum-local reward approach to ensure that performance hampering agents are not created. We evaluate the system using a Python-based simulator that has been designed over multiple network topologies to show the effectiveness of the algorithm.</p>
</div>
<a class="ext_link" href="https://ieeexplore.ieee.org/document/10041337" target="_blank"><i class="fa fa-external-link"></i></a>
<div style="clear:both"></div>
</div>
<br>
<!-- publication item -->
<div class="publication_item" data-groups='["all","CONFERENCES"]' data-date-publication="2022-03-13">
<div class="media">
<a href=".publication-detail23" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>DeepDish on a diet: Low-Latency, Energy-Efficient Object-Detection and Tracking at the Edge</h3>
<h4></h4>
<span class="publication_description">Proceedings of the 5th International Workshop on Edge Systems, Analytics and Networking</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Conferences</span>
<span class="label pdf"><a href="https://dl.acm.org/doi/10.1145/3517206.3526273" target="_blank"><font weight='bold' color="white">PDF</font></a></span>
<span class="publication_authors">Matthew Danish, <b>Rohit Verma</b>, Justas Brazauskas, Ian Lewis and Richard Mortier</span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail23 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">DeepDish on a diet: Low-Latency, Energy-Efficient Object-Detection and Tracking at the Edge</h3>
<span class="publication_authors">Matthew Danish, <b>Rohit Verma</b>, Justas Brazauskas, Ian Lewis and Richard Mortier</span>
<span class="label label-success">Conferences</span>
<p class="project_desc">Intelligent sensors using deep learning to comprehend video streams have become commonly used to track and analyse the movement of people and vehicles in public spaces. The models and hardware become more powerful at regular and frequent intervals. However, this computational marvel has come at the expense of heavy energy usage. If intelligent sensors are to become ubiquitous, such as being installed at every junction and frequently along every street in a city, then their power draw will become non-trivial, posing a severe downside to their usage. We explore Multi-Object Tracking (MOT) solutions based on our custom system that use less power while still maintaining reasonable accuracy.</p>
</div>
<a class="ext_link" href="https://dl.acm.org/doi/10.1145/3517206.3526273" target="_blank"><i class="fa fa-external-link"></i></a>
<div style="clear:both"></div>
</div>
<br>
<div class="publication_item" data-groups='["all","JOURNAL PAPERS"]' data-date-publication="2022-02-28">
<div class="media">
<a href=".publication-detail22" class="ex-link open_popup" data-effect="mfp-zoom-out"><i class="fa fa-plus-square-o"></i></a>
<div class="media-body">
<h3>Impact of Driving Behavior on Commuter’s Comfort during Cab Rides: Towards a New Perspective of Driver Rating</h3>
<h4></h4>
<span class="publication_description">ACM Transactions on Intelligent Systems and Technology</span> </div>
<hr style="margin:8px auto">
<span class="label label-success">Journal Paper</span>
<span class="label pdf"><a href="https://dl.acm.org/doi/10.1145/3523063" target="_blank"><font weight='bold' color="white">PDF</font></a></span>
<span class="publication_authors"><strong>Rohit Verma</strong>, Sugandh Pargal, Debasree Das, Tanushree Parbat, S. Sai Kambalapalli, Bivas Mitra, Sandip Chakraborty</span>
</div>
<div class="mfp-hide mfp-with-anim publication-detail22 publication-detail">
<!-- <div class="image_work">
<img class="img-responsive" src="" alt="img" width="480" height="200">
</div> -->
<div class="project_content">
<h3 class="publication_title">Impact of Driving Behavior on Commuter’s Comfort during Cab Rides: Towards a New Perspective of Driver Rating</h3>
<span class="publication_authors"><strong>Rohit Verma</strong>, Sugandh Pargal, Debasree Das, Tanushree Parbat, S. Sai Kambalapalli, Bivas Mitra, Sandip Chakraborty</span>
<span class="label label-success">Journal Paper</span>
<p class="project_desc">Commuter comfort in cab rides affects driver rating as well as the reputation of ride-hailing firms like Uber/Lyft. Existing research has revealed that commuter comfort not only varies at a personalized level but also is perceived differently on different trips for the same commuter. Furthermore, there are several factors, including driving behavior and driving environment, affecting the perception of comfort. Automatically extracting the perceived comfort level of a commuter due to the impact of the driving behavior is crucial for a timely feedback to the drivers, which can help them to meet the commuter's satisfaction. In light of this, we surveyed around 200 commuters who usually take such cab rides and obtained a set of features that impact comfort during cab rides. Following this, we develop a system Ridergo which collects smartphone sensor data from a commuter, extracts the spatial time series feature from the data, and then computes the level of commuter comfort on a five-point scale with respect to the driving. Ridergo uses a Hierarchical Temporal Memory model-based approach to observe anomalies in the feature distribution and then trains a Multi-task learning-based neural network model to obtain the comfort level of the commuter at a personalized level. The model also intelligently queries the commuter to add new data points to the available dataset and, in turn, improve itself over periodic training. Evaluation of Ridergo on 30 participants shows that the system could provide efficient comfort score with high accuracy when the driving impacts the perceived comfort.</p>
</div>
<a class="ext_link" href="https://dl.acm.org/doi/10.1145/3523063" target="_blank"><i class="fa fa-external-link"></i></a>
<div style="clear:both"></div>
</div>