From Machine Learning to Machine Reasoning

TL;DR

The paper proposes a unified view of reasoning as algebraic manipulation of knowledge, bridging simple learning modules and complex inference systems to build reasoning capabilities from basic components.

Contribution

It introduces a conceptual framework that treats various reasoning processes as algebraic operations on learned models, emphasizing a continuum between simple manipulations and sophisticated inference.

Findings

Shows that concatenating trained modules can be viewed as algebraic operations.

Suggests building reasoning capabilities by enriching algebraic manipulations of models.

Proposes a continuum between simple manipulations and complex inference systems.

Abstract

A plausible definition of "reasoning" could be "algebraically manipulating previously acquired knowledge in order to answer a new question". This definition covers first-order logical inference or probabilistic inference. It also includes much simpler manipulations commonly used to build large learning systems. For instance, we can build an optical character recognition system by first training a character segmenter, an isolated character recognizer, and a language model, using appropriate labeled training sets. Adequately concatenating these modules and fine tuning the resulting system can be viewed as an algebraic operation in a space of models. The resulting model answers a new question, that is, converting the image of a text page into a computer readable text. This observation suggests a conceptual continuity between algebraically rich inference systems, such as logical or probabilistic inference, and simple manipulations, such as the mere concatenation of trainable learning systems. Therefore, instead of trying to bridge the gap between machine learning systems and sophisticated "all-purpose" inference mechanisms, we can instead algebraically enrich the set of manipulations applicable to training systems, and build reasoning capabilities from the ground up.