SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.08924 · 2021

Epistemic Neural Networks

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 7 functions out of this paper's own repositories and ran 7 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
google-deepmind/enn pwc_unofficial 7 of 7
FunctionStatusWhere it lives
binary_log_likelihood Ran google-deepmind/enn/enn/losses/vae_losses.py
code served (permissive licence) · get_code("e4d50b188a65c5ad")
gaussian_log_likelihood Ran google-deepmind/enn/enn/losses/vae_losses.py
code served (permissive licence) · get_code("cf2414e3657a07d7")
l2_weights_with_predicate Ran google-deepmind/enn/enn/losses/utils.py
code served (permissive licence) · get_code("1c0c6f90dd25e0cc")
latent_kl_divergence Ran google-deepmind/enn/enn/losses/vae_losses.py
code served (permissive licence) · get_code("d72946fcf12cdc6a")
make_default_logger Ran google-deepmind/enn/enn/loggers.py
code served (permissive licence) · get_code("e30e012ccc8f6abb")
serialize Ran google-deepmind/enn/enn/loggers.py
code served (permissive licence) · get_code("73a9d7f8f0ed39a0")
variance_kl Ran google-deepmind/enn/enn/losses/prior_losses.py
code served (permissive licence) · get_code("be5ccac1b0191bdf")

Repositories linked to this paper

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

Abstract

Intelligence relies on an agent's knowledge of what it does not know. This capability can be assessed based on the quality of joint predictions of labels across multiple inputs. In principle, ensemble-based approaches produce effective joint predictions, but the computational costs of training large ensembles can become prohibitive. We introduce the epinet: an architecture that can supplement any conventional neural network, including large pretrained models, and can be trained with modest incremental computation to estimate uncertainty. With an epinet, conventional neural networks outperform very large ensembles, consisting of hundreds or more particles, with orders of magnitude less computation. The epinet does not fit the traditional framework of Bayesian neural networks. To accommodate development of approaches beyond BNNs, such as the epinet, we introduce the epistemic neural network (ENN) as an interface for models that produce joint predictions.

For agents

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

get_harvested_code_for_paper("2107.08924")
get_code_for_paper("2107.08924")
have("2107.08924")

Connect an agent — have() is free.