Learning Simple Algorithms from Examples
Wojciech Zaremba, Tomáš Mikolov, Armand Joulin et al. · arXiv (Cornell University) · 2015
We present an approach for learning simple algorithms such as copying, multi-digit addition and single digit multiplication directly from examples. Our framework consists of a set of interfaces, accessed by a controller. Typical interfaces are 1-D tapes or 2-D grids that hold the input and output data. For the controller, we explore a range of neural network-based models which vary in their ability to abstract the underlying algorithm from training instances and generalize to test examples with many thousands of digits. The controller is trained using $Q$-learning with several enhancements and we show that the bottleneck is in the capabilities of the controller rather than in the search incurred by $Q$-learning.
Cited by 22 works. doi:10.48550/arxiv.1511.07275 Open-access PDF
Authors
References
- Long Short-Term Memory
- Adaptation in Natural and Artificial Systems
- Genetic Algorithms in Search, Optimization and Machine Learning
- Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
- Learning from Delayed Rewards
- A formal theory of inductive inference. Part II
- Playing Atari with Deep Reinforcement Learning
- Learning Dependency-Based Compositional Semantics
- Optimal Ordered Problem Solver
- Reinforcement Learning Neural Turing Machines - Revised