Probabilistic least squares approach to ordinal regression

Srijith, PK and Shevade, Shirish and Sundararajan, S (2012) Probabilistic least squares approach to ordinal regression. In: Proceedings of the 25th Australasian Joint Conference, December 4-7, 2012, Sydney, Australia.

PDF lncs_7691_683_2012.pdf - Published Version Restricted to Registered users only Download (320kB) | Request a copy

Abstract

This paper proposes a novel approach to solve the ordinal regression problem using Gaussian processes. The proposed approach, probabilistic least squares ordinal regression (PLSOR), obtains the probability distribution over ordinal labels using a particular likelihood function. It performs model selection (hyperparameter optimization) using the leave-one-out cross-validation (LOO-CV) technique. PLSOR has conceptual simplicity and ease of implementation of least squares approach. Unlike the existing Gaussian process ordinal regression (GPOR) approaches, PLSOR does not use any approximation techniques for inference. We compare the proposed approach with the state-of-the-art GPOR approaches on some synthetic and benchmark data sets. Experimental results show the competitiveness of the proposed approach.

Item Type:	Conference Proceedings
Publisher:	Springer
Additional Information:	Copyright of this article belongs to Springer.
Keywords:	Gaussian Processes; Ordinal Regression; Probabilistic Least; Squares; Cross-Validation
Department/Centre:	Division of Electrical Sciences > Computer Science & Automation
Date Deposited:	20 Nov 2013 11:46
Last Modified:	20 Nov 2013 11:46
URI:	http://eprints.iisc.ac.in/id/eprint/47812

Actions (login required)

View Item