DeepAI AI Chat
Log In Sign Up

End-to-end Algorithm Synthesis with Recurrent Networks: Logical Extrapolation Without Overthinking

by   Arpit Bansal, et al.

Machine learning systems perform well on pattern matching tasks, but their ability to perform algorithmic or logical reasoning is not well understood. One important reasoning capability is logical extrapolation, in which models trained only on small/simple reasoning problems can synthesize complex algorithms that scale up to large/complex problems at test time. Logical extrapolation can be achieved through recurrent systems, which can be iterated many times to solve difficult reasoning problems. We observe that this approach fails to scale to highly complex problems because behavior degenerates when many iterations are applied – an issue we refer to as "overthinking." We propose a recall architecture that keeps an explicit copy of the problem instance in memory so that it cannot be forgotten. We also employ a progressive training routine that prevents the model from learning behaviors that are specific to iteration number and instead pushes it to learn behaviors that can be repeated indefinitely. These innovations prevent the overthinking problem, and enable recurrent systems to solve extremely hard logical extrapolation tasks, some requiring over 100K convolutional layers, without overthinking.


page 1

page 14

page 15


Thinking Deeply with Recurrence: Generalizing from Easy to Hard Sequential Reasoning Problems

Deep neural networks are powerful machines for visual pattern recognitio...

SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver

Integrating logical reasoning within deep learning architectures has bee...

Can You Learn an Algorithm? Generalizing from Easy to Hard Problems with Recurrent Networks

Deep neural networks are powerful machines for visual pattern recognitio...

On the Paradox of Learning to Reason from Data

Logical reasoning is needed in a wide range of NLP tasks. Can a BERT mod...

From Machine Learning to Machine Reasoning

A plausible definition of "reasoning" could be "algebraically manipulati...

Tunneling Neural Perception and Logic Reasoning through Abductive Learning

Perception and reasoning are basic human abilities that are seamlessly c...

Logical Tasks for Measuring Extrapolation and Rule Comprehension

Logical reasoning is essential in a variety of human activities. A repre...