Pith. sign in

REVIEW 5 cited by

Epistemic Neural Networks

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2107.08924 v8 pith:RTX75NVR submitted 2021-07-19 cs.LG cs.AIstat.ML

Ian Osband , Zheng Wen , Seyed Mohammad Asghari , Vikranth Dwaracherla , Morteza Ibrahimi , Xiuyuan Lu , Benjamin Van Roy This is my paper
classification cs.LGcs.AIstat.ML
keywords neuralepinetjointlargenetworkspredictionsapproachescomputation
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original 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.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A Mutual Information Lower Bound for Multimodal Regression Active Learning

    cs.LG 2026-05 unverdicted novelty 7.0 of 10

    Derives MI-LB acquisition function from mutual information in a two-index epistemic-aleatoric framework and shows it outperforms baselines on multimodal regression benchmarks.

  2. Benchmarking and Improving Monitors for Out-Of-Distribution Alignment Failure in LLMs

    cs.AI 2026-05 unverdicted novelty 6.0 of 10

    MOOD benchmark shows guard models fail to generalize to OOD alignment failures in LLMs, but combining them with Mahalanobis and perplexity OOD detectors improves recall from 39% to 45% with better scaling than larger ...

  3. Benchmarking and Improving Monitors for Out-Of-Distribution Alignment Failure in LLMs

    cs.AI 2026-05 conditional novelty 6.0 of 10

    Introduces MOOD benchmark for OOD LLM alignment failures and shows guard models plus Mahalanobis and perplexity OOD detectors improve recall from 39% to 45% with positive scaling.

  4. Structurally Separated Uncertainty in Supervised Latent Variable Models

    cs.LG 2026-02 conditional novelty 6.0 of 10

    A credal concept-bottleneck model that supervises aleatoric uncertainty with annotator disagreement and epistemic uncertainty with prediction error yields near-zero correlation between the two uncertainty estimates.

  5. Composite Bayesian Optimization In Function Spaces Using NEON -- Neural Epistemic Operator Networks

    cs.LG 2024-04 unverdicted novelty 6.0 of 10

    NEON provides uncertainty-aware operator learning for composite Bayesian optimization in function spaces using a single network, achieving claimed SOTA with orders of magnitude fewer parameters than ensembles.

Pith tools