qqn-optimizer/CITATION.cff at master · SimiaCryptus/qqn-optimizer · GitHub

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
# CITATION.cff
# This file provides citation information for the QQN Optimizer software
# See https://citation-file-format.github.io/ for format specification

cff-version: 1.0.0
message: "If you use this software, please cite it as below."
type: software
title: "QQN: A Quadratic Hybridization of Quasi-Newton Methods for Nonlinear Optimization"
abstract: >
  We present the Quadratic-Quasi-Newton (QQN) algorithm, which combines gradient and
  quasi-Newton directions through quadratic interpolation. QQN constructs a parametric
  path d(t) = t(1-t)(-∇f) + t²d_L-BFGS and performs univariate optimization along this
  path, creating an adaptive interpolation that requires no additional hyperparameters
  beyond those of its constituent methods. Our comprehensive evaluation across 62
  benchmark problems demonstrates QQN's superior robustness and practical utility with
  statistical dominance over L-BFGS and Adam variants.

authors:
  - family-names: "Charneski"
    given-names: "Andrew"
    affiliation: "SimiaCryptus Software"
    orcid: "https://orcid.org/0000-0000-0000-0000"  # Add if available

repository-code: "https://github.com/SimiaCryptus/qqn-optimizer"
url: "https://simiacryptus.github.io/qqn-optimizer"
license: MIT
version: "1.0.0"
date-released: "2025-08-02"

keywords:
  - optimization
  - quasi-newton
  - quadratic-interpolation
  - l-bfgs
  - gradient-descent
  - numerical-methods
  - machine-learning
  - rust
  - research-software
  - benchmarking
  - nonlinear-optimization
  - statistical-analysis

references:
  - type: article
    title: "QQN: A Quadratic Hybridization of Quasi-Newton Methods for Nonlinear Optimization"
    authors:
      - family-names: "Charneski"
        given-names: "Andrew"
        affiliation: "SimiaCryptus Software"
    journal: "arXiv preprint"
    year: 2025
    month: 8
    url: "https://github.com/SimiaCryptus/qqn-optimizer/"
    abstract: >
      We present the Quadratic-Quasi-Newton (QQN) algorithm, which combines gradient and
      quasi-Newton directions through quadratic interpolation. QQN constructs a parametric
      path and performs univariate optimization along this path, creating an adaptive
      interpolation that requires no additional hyperparameters beyond those of its
      constituent methods.

  - type: article
    title: "On the limited memory BFGS method for large scale optimization"
    authors:
      - family-names: "Liu"
        given-names: "Dong C."
      - family-names: "Nocedal"
        given-names: "Jorge"
    journal: "Mathematical Programming"
    volume: 45
    issue: "1-3"
    start: 503
    end: 528
    year: 1989
    doi: "10.1007/BF01589116"

  - type: book
    title: "Numerical Optimization"
    authors:
      - family-names: "Nocedal"
        given-names: "Jorge"
      - family-names: "Wright"
        given-names: "Stephen J."
    publisher: "Springer Science & Business Media"
    year: 2006
    edition: "2nd"
    isbn: "978-0-387-30303-1"

  - type: software
    title: "COCO: A platform for comparing continuous optimizers in a black-box setting"
    authors:
      - family-names: "Hansen"
        given-names: "Nikolaus"
      - family-names: "Auger"
        given-names: "Anne"
      - family-names: "Ros"
        given-names: "Raymond"
      - family-names: "Mersmann"
        given-names: "Olaf"
      - family-names: "Tušar"
        given-names: "Tea"
      - family-names: "Brockhoff"
        given-names: "Dimo"
    year: 2016
    doi: "10.48550/arXiv.1603.08785"

preferred-citation:
  type: article
  title: "QQN: A Quadratic Hybridization of Quasi-Newton Methods for Nonlinear Optimization"
  authors:
    - family-names: "Charneski"
      given-names: "Andrew"
      affiliation: "SimiaCryptus Software"
  journal: "arXiv preprint"
  year: 2025
  month: 8
  url: "https://github.com/SimiaCryptus/qqn-optimizer/"
  repository-code: "https://github.com/SimiaCryptus/qqn-optimizer"
  abstract: >
    We present the Quadratic-Quasi-Newton (QQN) algorithm, which combines gradient and
    quasi-Newton directions through quadratic interpolation. QQN constructs a parametric
    path d(t) = t(1-t)(-∇f) + t²d_L-BFGS and performs univariate optimization along this
    path, creating an adaptive interpolation that requires no additional hyperparameters
    beyond those of its constituent methods. Our comprehensive evaluation across 62
    benchmark problems demonstrates QQN's superior robustness and practical utility.
  keywords:
    - optimization
    - quasi-Newton methods
    - L-BFGS
    - gradient descent
    - quadratic interpolation
    - benchmarking
    - statistical analysis