LGLOMLJun 10, 2020

A Probabilistic Model for Discriminative and Neuro-Symbolic Semi-Supervised Learning

arXiv:2006.05896v45 citations
AI Analysis

This work addresses semi-supervised learning for machine learning practitioners, offering a unified framework that integrates discriminative and neuro-symbolic methods, but it appears incremental as it builds upon and refines existing approaches.

The paper tackles the problem of semi-supervised learning by presenting a probabilistic model for discriminative SSL that mirrors generative approaches, showing that existing methods can be interpreted as approximations and improved upon, and extends it to neuro-symbolic SSL by linking logical rules to the model's prior, unifying statistical learning with logical reasoning.

Much progress has been made in semi-supervised learning (SSL) by combining methods that exploit different aspects of the data distribution, e.g. consistency regularisation relies on properties of $p(x)$, whereas entropy minimisation pertains to the label distribution $p(y|x)$. Focusing on the latter, we present a probabilistic model for discriminative SSL, that mirrors its classical generative counterpart. Under the assumption $y|x$ is deterministic, the prior over latent variables becomes discrete. We show that several well-known SSL methods can be interpreted as approximating this prior, and can be improved upon. We extend the discriminative model to neuro-symbolic SSL, where label features satisfy logical rules, by showing such rules relate directly to the above prior, thus justifying a family of methods that link statistical learning and logical reasoning, and unifying them with regular SSL.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes