Learning Simple Algorithms from Examples

Abstract

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.

References in corpus (2)

• Playing Atari with Deep Reinforcement Learning
• Reinforcement Learning Neural Turing Machines - Revised

Cited by in corpus (17)

• Neural Programmer-Interpreters
• An Actor-Critic Algorithm for Sequence Prediction
• Learning to Optimize
• Model-Predictive Policy Learning with Uncertainty Regularization for Driving in Dense Traffic
• MazeBase: A Sandbox for Learning from Games
• Dynamic Neural Turing Machine with Soft and Hard Addressing Schemes
• Adaptive Neural Compilation
• Learning Efficient Algorithms with Hierarchical Attentive Memory
• Improving the Neural GPU Architecture for Algorithm Learning
• Extensions and Limitations of the Neural GPU
• Learning to Learn from Noisy Web Videos
• Disentangled Representations in Neural Models
• ACtuAL: Actor-Critic Under Adversarial Learning
• Differentiable Functional Program Interpreters
• Common-Description Learning: A Framework for Learning Algorithms and Generating Subproblems from Few Examples
• Neural Program Meta-Induction
• Neural Programming by Example