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
|
@article{dropout,
author = {Srivastava, Nitish and Hinton, Geoffrey and Krizhevsky, Alex and Sutskever, Ilya and Salakhutdinov, Ruslan},
title = {Dropout: A Simple Way to Prevent Neural Networks from Overfitting},
journal = {Journal of Machine Learning Research},
issue_date = {January 2014},
volume = {15},
number = {1},
month = January,
year = {2014},
issn = {1532-4435},
pages = {1929--1958},
numpages = {30},
url = {http://dl.acm.org/citation.cfm?id=2627435.2670313},
acmid = {2670313},
publisher = {JMLR.org},
keywords = {deep learning, model combination, neural networks, regularization},
}
@manual{cudnn,
title = {cuDNN Library},
author = {Nvidia},
edition = {5.1},
month = {May},
year = {2016}
}
@INCOLLECTION{
Kearfott1996ADo,
author = "R. Baker Kearfott",
editor = "Martin Berz and Christian Bischof and George Corliss and Andreas Griewank",
title = "Automatic Differentiation of Conditional Branches in an Operator Overloading Context",
booktitle = "Computational Differentiation: Techniques, Applications, and Tools",
pages = "75--81",
publisher = "SIAM",
address = "Philadelphia, PA",
key = "Kearfott1996ADo",
crossref = "Berz1996CDT",
abstract = "In the past, it has been problematical to include {\tt IF-THEN-ELSE} branches
in automatic differentiation processes driven by operator overloading and code list generation, when
the branch condition contains variables. However, this problem can be circumvented with a special
``branch function'' $\chi$. Definition of this function, formulas for its use, and
implications of its use will be discussed. A second issue is: what can be done when derivatives are
discontinuous? In fact, simple and meaningful Newton iterations can be set up when even the function
itself is discontinuous. Simplified figures and examples are given, as well as references to
in-depth explanations. An example of the convergence behavior is given with an interval Newton
method to find critical points for the problem ``$\min |x|$.''",
keywords = "Conditional branches, operator overloading, branch function, discontinuous
derivatives.",
referred = "[Berz2002TaU], [Dignath2002AAa].",
year = "1996"
}
@INCOLLECTION{
Tadjouddine2005ItP,
author = "Mohamed Tadjouddine and Frances Bodman and John D. Pryce and Shaun A. Forth",
title = "Improving the Performance of the Vertex Elimination Algorithm for Derivative
Calculation",
editor = "H. M. B{\"u}cker and G. Corliss and P. Hovland and U. Naumann and B.
Norris",
booktitle = "Automatic Differentiation: {A}pplications, Theory, and Implementations",
series = "Lecture Notes in Computational Science and Engineering",
publisher = "Springer",
year = "2005",
abstract = "In previous work [TOMS, 2004, 30(3), 266--299], we used Markowitz-like heuristics
to find elimination sequences that minimise the number of floating-point operations (flops) for
vertex elimination Jacobian code. We also used the depth-first traversal algorithm to reorder the
statements of the Jacobian code with the aim of reducing the number of memory accesses. In this
work, we study the effects of reducing flops or memory accesses within the vertex elimination
algorithm for Jacobian calculation. On RISC processors, we observed that for data residing in
registers, the number of flops gives a good estimate of the execution time, while for
out-of-register data, the execution time is dominated by the time for memory access operations. We
also present a statement reordering scheme based on a greedy list scheduling algorithm using ranking
functions. This statement reordering will enable us to trade off the exploitation of the instruction
level parallelism of such processors with the reduction in memory accesses.",
crossref = "Bucker2005ADA",
ad_tools = "EliAD",
ad_theotech = "X-Country",
pages = "111--120",
doi = "10.1007/3-540-28438-9_10"
}
@ARTICLE{
Tadjouddine2008VoA,
author = "Tadjouddine, E. M.",
title = "{Vertex-ordering Algorithms for Automatic Differentiation of Computer Codes}",
journal = "The Computer Journal",
volume = "51",
number = "6",
pages = "688--699",
doi = "10.1093/comjnl/bxm115",
year = "2008",
abstract = "In the context of Automatic Differentiation (AD) of functions represented by
computer code via the vertex elimination approach first advocated by Griewank and Reese (On the
Calculation of Jacobian Matrices by the Markowitz Rule. In Griewank, A. and Corliss, G.F. (eds),
Automatic Differentiation of Algorithms: Theory, Implementation and Application, pp. 126-135. SIAM,
1991, Philadelphia, PA.), we present two approximate algorithms based on the linearized
computational graph of the input code. The first is a statement-reordering algorithm aiming to tune
the AD-generated code so as to maximize its performance for modern superscalar processors. The
second is aimed at detecting interface contractions introduced by Bischof and Haghighat
(Hierarchical Approaches to Automatic Differentiation. In Berz, M., Bischof, C., Corliss, G. and
Griewank, A. (eds), Computational Differentiation: Techniques, Applications, and Tools, pp. 83-94.
SIAM, 1996, Philadelphia, PA) in order to enable exploitation of the structure of the input code in
the differentiation process. Performance data are also presented.",
url = "http://comjnl.oxfordjournals.org/cgi/content/abstract/51/6/688",
eprint = "http://comjnl.oxfordjournals.org/cgi/reprint/51/6/688.pdf",
ad_theotech = "Hierarchical Approach"
}
@INCOLLECTION{
Griewank1991OtC,
author = "Andreas Griewank and Shawn Reese",
editor = "Andreas Griewank and George F. Corliss",
title = "On the Calculation of {J}acobian Matrices by the {M}arkowitz Rule",
booktitle = "Automatic Differentiation of Algorithms: Theory, Implementation, and Application",
pages = "126--135",
publisher = "SIAM",
address = "Philadelphia, PA",
key = "Griewank1991OtC",
crossref = "Griewank1991ADo",
comment = "Also appeared as Preprint MCS--P267--1091, Mathematics and Computer Science
Division, Argonne National Laboratory, Argonne, Ill., January 1992.",
referred = "[Bischof1996HAt], [Corl91a]; [Feehery1996ADB], [Irim91a], [Naumann2002ETf],
[Tadjouddine2001ATa].",
isbn = "0--89871--284--X",
year = "1991",
ad_theotech = "X-Country"
}
|
