Model-Free Episodic Control
Charles Blundell, Benigno Uría, Alexander Pritzel et al. · arXiv (Cornell University) · 2016
State of the art deep reinforcement learning algorithms take many millions of interactions to attain human-level performance. Humans, on the other hand, can very quickly exploit highly rewarding nuances of an environment upon first discovery. In the brain, such rapid learning is thought to depend on the hippocampus and its capacity for episodic memory. Here we investigate whether a simple model of hippocampal episodic control can learn to solve difficult sequential decision-making tasks. We demonstrate that it not only attains a highly rewarding strategy significantly faster than state-of-the-art deep reinforcement learning algorithms, but also achieves a higher overall reward on some of the more challenging domains.
Cited by 152 works. doi:10.48550/arxiv.1606.04460 Open-access PDF
Authors
- Charles Blundell
- Benigno Uría
- Alexander Pritzel
- Yazhe Li
- Avraham Ruderman
- Joel Z. Leibo
- Jack W. Rae
- Daan Wierstra
- Demis Hassabis
References
- Gradient-based learning applied to document recognition
- Neural networks and physical systems with emergent collective computational abilities.
- Human-level control through deep reinforcement learning
- Mastering the game of Go with deep neural networks and tree search
- Information theory, inference, and learning algorithms
- Learning to generate chairs with convolutional neural networks
- Place navigation impaired in rats with hippocampal lesions
- Memory and the hippocampus: A synthesis from findings with rats, monkeys, and humans.
- Stochastic Backpropagation and Approximate Inference in Deep Generative Models
- Why there are complementary learning systems in the hippocampus and neocortex: Insights from the successes and failures of connectionist models of learning and memory.