Successor Feature Neural Episodic Control
Not on the map. Why?
Abstract (arXiv)
A longstanding goal in reinforcement learning is to build intelligent agents that show fast learning and a flexible transfer of skills akin to humans and animals. This paper investigates the integration of two frameworks for tackling those goals: episodic control and successor features. Episodic control is a cognitively inspired approach relying on episodic memory, an instance-based memory model of an agent's experiences. Meanwhile, successor features and generalized policy improvement (SF&GPI) is a meta and transfer learning framework allowing to learn policies for tasks that can be efficiently reused for later tasks which have a different reward function. Individually, these two techniques have shown impressive results in vastly improving sample efficiency and the elegant reuse of previously learned policies. Thus, we outline a combination of both approaches in a single reinforcement learning framework and empirically illustrate its benefits.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 22 of its 48 references that are in the map. First posted on arXiv 4 Nov 2021.
Builds on
22 of 48 references.
References · 22
- 2013Playing Atari with Deep Reinforcement Learningcited by 3,211
- 2017Billion-scale similarity search with GPUscited by 2,477
- 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,327
- 2019Dota 2 with Large Scale Deep Reinforcement Learningcited by 1,084
- 2020Meta-Learning in Neural Networks: A Surveycited by 944
- 2017A Brief Survey of Deep Reinforcement Learningcited by 466
- 2016Successor Features for Transfer in Reinforcement Learningcited by 331
- 2020Transfer Learning in Deep Reinforcement Learning: A Surveycited by 259
- 2017Neural Episodic Controlcited by 148
- 2019Fast Task Inference with Variational Intrinsic Successor Featurescited by 131
- 2018Universal Successor Features Approximatorscited by 102
- 2016Model-Free Episodic Controlcited by 100
10 more
- 2019Transfer in Deep Reinforcement Learning Using Successor Features and Generalised Policy Improvementcited by 88
- 2018Episodic Memory Deep Q-Networkscited by 41
- 2020Deep Reinforcement Learning and its Neuroscientific Implicationscited by 33
- 2021Generalizable Episodic Memory for Deep Reinforcement Learningcited by 22
- 2020Universal Successor Features for Transfer Reinforcement Learningcited by 20
- 2021Successor Feature Sets: Generalizing Successor Representations Across Policiescited by 6
- 2019Better transfer learning with inferred successor mapscited by 4
- 2019Memory-Efficient Episodic Control Reinforcement Learning with Dynamic Online k-meanscited by 3
- 2019State2vec: Off-Policy Successor Features Approximatorscited by 2
- 2019Sample-Efficient Reinforcement Learning with Maximum Entropy Mellowmax Episodic Controlcited by 1
Built on by
Citing papers per year: 2023: 2; 2024: 1.
3 papers.