Computer Science > Machine Learning
[Submitted on 19 Jul 2021 (v1), revised 6 Jul 2022 (this version, v5), latest version 17 May 2023 (v8)]
Title:Epistemic Neural Networks
View PDFAbstract: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. Conventional neural networks lack this capability and, since most research has focused on marginal predictions, this shortcoming has been largely overlooked. We introduce the epistemic neural network (ENN) as an interface for models that represent uncertainty as required to generate useful joint predictions. While prior approaches to uncertainty modeling such as Bayesian neural networks can be expressed as ENNs, this new interface facilitates comparison of joint predictions and the design of novel architectures and algorithms. In particular, 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. We demonstrate this efficacy across synthetic data, ImageNet, and some reinforcement learning tasks. As part of this effort we open-source experiment code.
Submission history
From: Ian Osband [view email][v1] Mon, 19 Jul 2021 14:37:57 UTC (3,591 KB)
[v2] Wed, 13 Apr 2022 16:05:40 UTC (7,910 KB)
[v3] Wed, 25 May 2022 22:37:30 UTC (14,443 KB)
[v4] Mon, 30 May 2022 15:00:45 UTC (14,448 KB)
[v5] Wed, 6 Jul 2022 12:41:37 UTC (14,860 KB)
[v6] Wed, 5 Oct 2022 23:17:42 UTC (6,085 KB)
[v7] Thu, 2 Feb 2023 15:53:20 UTC (6,522 KB)
[v8] Wed, 17 May 2023 21:17:29 UTC (6,468 KB)
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
IArxiv Recommender
(What is IArxiv?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.