% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@MASTERSTHESIS{Fischer:888741,
author = {Fischer, Kirsten},
othercontributors = {Helias, Moritz and Dahmen, David},
title = {{D}ecomposition of {D}eep {N}eural {N}etworks into
{C}orrelation {F}unctions},
school = {RWTH Aachen University},
type = {Masterarbeit},
reportid = {FZJ-2020-05175},
pages = {91 p.},
year = {2020},
note = {Masterarbeit, RWTH Aachen University, 2020},
abstract = {Recent years have shown a great success of deep neural
networks. One active field of research investigates the
functioning mechanisms of such networks with respect to the
network expressivity as well as information processing
within the network. In this thesis, we describe the
input-output mapping implemented by deep neural networks in
terms of correlation functions. To trace the transformation
of correlation functions within neural networks, we make use
of methods from statistical physics. Using a quadratic
approximation for non-linear activation functions, we obtain
recursive relations in a perturbative manner by means of
Feynman diagrams. Our results yield a characterization of
the network as a non-linear mapping of mean and covariance,
which can be extended by including corrections from higher
order correlations. Furthermore, re-expressing the training
objective in terms of data correlations allows us to study
their role for solutions to a given task. First, we
investigate an adaptation of the XOR problem, in which case
the solutions implemented by neural networks can largely be
described in terms of mean and covariance of each class.
Furthermore, we study the MNIST database as an example of a
non-synthetic dataset. For MNIST, solutions based on
empirical estimates for mean and covariance of each class
already capture a large amount of the variability within the
dataset, but still exhibit a non-negligible performance gap
in comparison to solutions based on the actual dataset.
Lastly, we introduce an example task where higher order
correlations exclusively encode class membership, which
allows us to explore their role for solutions found by
neural networks. Finally, our framework also allows us to
make predictions regarding the correlation functions that
are inferable from data, yielding insights into the network
expressivity. This work thereby creates a link between
statistical physics and machine learning, aiming towards
explainable AI.},
cin = {INM-6 / IAS-6 / INM-10},
cid = {I:(DE-Juel1)INM-6-20090406 / I:(DE-Juel1)IAS-6-20130828 /
I:(DE-Juel1)INM-10-20170113},
pnm = {574 - Theory, modelling and simulation (POF3-574) /
RenormalizedFlows - Transparent Deep Learning with
Renormalized Flows (BMBF-01IS19077A) / MSNN - Theory of
multi-scale neuronal networks (HGF-SMHB-2014-2018) /
neuroIC002 - Recurrence and stochasticity for neuro-inspired
computation (EXS-SF-neuroIC002)},
pid = {G:(DE-HGF)POF3-574 / G:(DE-Juel-1)BMBF-01IS19077A /
G:(DE-Juel1)HGF-SMHB-2014-2018 / G:(DE-82)EXS-SF-neuroIC002},
typ = {PUB:(DE-HGF)19},
url = {https://juser.fz-juelich.de/record/888741},
}