Compressed Inference for Probabilistic Sequential Models

TL;DR

This paper introduces a novel polynomial time inference algorithm for probabilistic sequential models that directly estimates functional outcomes, improving efficiency and accuracy in applications like robot tracking and handwriting recognition.

Contribution

The paper proposes a new inference method for sequential models that directly computes outcomes, bypassing full state sequence estimation, with demonstrated benefits in multiple applications.

Findings

Outperforms traditional inference methods in accuracy.

Operates in polynomial time, making it computationally efficient.

Applicable to diverse tasks like robot tracking and handwriting recognition.

Abstract

Hidden Markov models (HMMs) and conditional random fields (CRFs) are two popular techniques for modeling sequential data. Inference algorithms designed over CRFs and HMMs allow estimation of the state sequence given the observations. In several applications, estimation of the state sequence is not the end goal; instead the goal is to compute some function of it. In such scenarios, estimating the state sequence by conventional inference techniques, followed by computing the functional mapping from the estimate is not necessarily optimal. A more formal approach is to directly infer the final outcome from the observations. In particular, we consider the specific instantiation of the problem where the goal is to find the state trajectories without exact transition points and derive a novel polynomial time inference algorithm that outperforms vanilla inference techniques. We show that this particular problem arises commonly in many disparate applications and present experiments on three of them: (1) Toy robot tracking; (2) Single stroke character recognition; (3) Handwritten word recognition.