Deep learning is a broad set of techniques that uses multiple layers of representation to automatically learn relevant features directly from structured data. Recently, such techniques have yielded record-breaking results on a diverse set of difficult machine learning tasks in computer vision, speech recognition, and natural language processing. Despite the enormous success of deep learning, relatively little is understood theoretically about why these techniques are so successful at feature learning and compression. Here, we show that deep learning is intimately related to one of the most important and successful techniques in theoretical physics, the renormalization group (RG). RG is an iterative coarse-graining scheme that allows for the extraction of relevant features (i.e. operators) as a physical system is examined at different length scales. We construct an exact mapping from the variational renormalization group, first introduced by Kadanoff, and deep learning architectures based on Restricted Boltzmann Machines (RBMs). We illustrate these ideas using the nearest-neighbor Ising Model in one and two-dimensions. Our results suggests that deep learning algorithms may be employing a generalized RG-like scheme to learn relevant features from data.
[1]
M. V. Rossum,et al.
In Neural Computation
,
2022
.
[2]
Michael I. Jordan,et al.
Advances in Neural Information Processing Systems 30
,
1995
.
[3]
J. Cardy.
Scaling and Renormalization in Statistical Physics
,
1996
.
[4]
Jason Weston,et al.
Large-scale kernel machines
,
2007
.
[5]
Zoubin Ghahramani,et al.
Proceedings of the 24th international conference on Machine learning
,
2007,
ICML 2007.
[6]
Peter A. Flach,et al.
Proceedings of the 28th International Conference on Machine Learning
,
2011
.
[7]
F. Bach,et al.
Optimization with Sparsity-Inducing Penalties (Foundations and Trends(R) in Machine Learning)
,
2011
.
[8]
Shay B. Cohen,et al.
Advances in Neural Information Processing Systems 25
,
2012,
NIPS 2012.
[9]
Pascal Vincent,et al.
Representation Learning: A Review and New Perspectives
,
2012,
IEEE Transactions on Pattern Analysis and Machine Intelligence.
[10]
Kathy P. Wheeler,et al.
Reviews of Modern Physics
,
2013
.
[11]
Physics Reports
,
2022
.