SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2209.04882 · 2022

A statistical mechanics framework for Bayesian deep neural networks beyond the infinite-width limit

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 8 functions out of this paper's own repositories and ran 8 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
rpacelli/fc_deep_bayesian_networks canonical 8 of 8
FunctionStatusWhere it lives
CorrMat Ran rpacelli/fc_deep_bayesian_networks/theory.py
code served (permissive licence) · get_code("67c39864469f28c6")
check_file Ran rpacelli/fc_deep_bayesian_networks/data_analysis/data_analysis.py
code served (permissive licence) · get_code("e633d84143b2409c")
find_device Ran rpacelli/fc_deep_bayesian_networks/utils.py
code served (permissive licence) · get_code("31c107ff46c82600")
k0 Ran rpacelli/fc_deep_bayesian_networks/theory.py
code served (permissive licence) · get_code("8292db5fe4044158")
kmatrix Ran rpacelli/fc_deep_bayesian_networks/theory.py
code served (permissive licence) · get_code("0e71a1474f7716e0")
load_run Ran rpacelli/fc_deep_bayesian_networks/data_analysis/data_analysis.py
code served (permissive licence) · get_code("9b91ed66d4895edf")
test Ran rpacelli/fc_deep_bayesian_networks/utils.py
code served (permissive licence) · get_code("76930061f3339640")
train Ran rpacelli/fc_deep_bayesian_networks/utils.py
code served (permissive licence) · get_code("3a5c5b3ec170bafa")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Despite the practical success of deep neural networks, a comprehensive theoretical framework that can predict practically relevant scores, such as the test accuracy, from knowledge of the training data is currently lacking. Huge simplifications arise in the infinite-width limit, where the number of units $N_\ell$ in each hidden layer ($\ell=1,\dots, L$, being $L$ the depth of the network) far exceeds the number $P$ of training examples. This idealisation, however, blatantly departs from the reality of deep learning practice. Here, we use the toolset of statistical mechanics to overcome these limitations and derive an approximate partition function for fully-connected deep neural architectures, which encodes information about the trained models. The computation holds in the ''thermodynamic limit'' where both $N_\ell$ and $P$ are large and their ratio $α_\ell = P/N_\ell$ is finite. This advance allows us to obtain (i) a closed formula for the generalisation error associated to a regression task in a one-hidden layer network with finite $α_1$; (ii) an approximate expression of the partition function for deep architectures (via an ''effective action'' that depends on a finite number of ''order parameters''); (iii) a link between deep neural networks in the proportional asymptotic limit and Student's $t$ processes.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2209.04882")
get_code_for_paper("2209.04882")
have("2209.04882")

Connect an agent — have() is free.