arXiv subjects

Marco Cox

Publications and source records attributed to Marco Cox.

4 recordsLinked to original sources

arXiv2021

On Preference Learning Based on Sequential Bayesian Optimization with Pairwise Comparison

User preference learning is generally a hard problem. Individual preferences are typically unknown even to users themselves, while the space of choices is infinite. Here we study user preference learning from information-theoretic perspective. We model preference learning as a system with two interacting sub-systems, one representing a user with his/her preferences and another one representing an agent that has to learn these preferences. The user with his/her behaviour is modeled by a parametric preference function. To efficiently learn the preferences and reduce search space quickly, we propose the agent that interacts with the user to collect the most informative data for learning. The agent presents two proposals to the user for evaluation, and the user rates them based on his/her preference function. We show that the optimum agent strategy for data collection and preference learning is a result of maximin optimization of the normalized weighted Kullback-Leibler (KL) divergence between true and agent-assigned predictive user response distributions. The resulting value of KL-divergence, which we also call remaining system uncertainty (RSU), provides an efficient performance metric in the absence of the ground truth. This metric characterises how well the agent can predict user and, thus, the quality of the underlying learned user (preference) model. Our proposed agent comprises sequential mechanisms for user model inference and proposal generation. To infer the user model (preference function), Bayesian approximate inference is used in the agent. The data collection strategy is to generate proposals, responses to which help resolving uncertainty associated with prediction of the user responses the most. The efficiency of our approach is validated by numerical simulations. Also a real-life example of preference learning application is provided.

cs.LG↗

arXiv2018

A Factor Graph Approach to Automated Design of Bayesian Signal Processing Algorithms

The benefits of automating design cycles for Bayesian inference-based algorithms are becoming increasingly recognized by the machine learning community. As a result, interest in probabilistic programming frameworks has much increased over the past few years. This paper explores a specific probabilistic programming paradigm, namely message passing in Forney-style factor graphs (FFGs), in the context of automated design of efficient Bayesian signal processing algorithms. To this end, we developed "ForneyLab" (https://github.com/biaslab/ForneyLab.jl) as a Julia toolbox for message passing-based inference in FFGs. We show by example how ForneyLab enables automatic derivation of Bayesian signal processing algorithms, including algorithms for parameter estimation and model comparison. Crucially, due to the modular makeup of the FFG framework, both the model specification and inference methods are readily extensible in ForneyLab. In order to test this framework, we compared variational message passing as implemented by ForneyLab with automatic differentiation variational inference (ADVI) and Monte Carlo methods as implemented by state-of-the-art tools "Edward" and "Stan". In terms of performance, extensibility and stability issues, ForneyLab appears to enjoy an edge relative to its competitors for automated inference in state-space models.

cs.LG↗

arXiv2018

A Probabilistic Modeling Approach to One-Shot Gesture Recognition

Gesture recognition enables a natural extension of the way we currently interact with devices. Commercially available gesture recognition systems are usually pre-trained and offer no option for customization by the user. In order to improve the user experience, it is desirable to allow end users to define their own gestures. This scenario requires learning from just a few training examples if we want to impose only a light training load on the user. To this end, we propose a gesture classifier based on a hierarchical probabilistic modeling approach. In this framework, high-level features that are shared among different gestures can be extracted from a large labeled data set, yielding a prior distribution for gestures. When learning new types of gestures, the learned shared prior reduces the number of required training examples for individual gestures. We implemented the proposed gesture classifier for a Myo sensor bracelet and show favorable results for the tested system on a database of 17 different gesture types. Furthermore, we propose and implement two methods to incorporate the gesture classifier in a real-time gesture recognition system.

eess.SP↗

arXiv2016

A Bayesian binary classification approach to pure tone audiometry

The pure tone hearing threshold is usually estimated from responses to stimuli at a set of standard frequencies. This paper describes a probabilistic approach to the estimation problem in which the hearing threshold is modelled as a smooth continuous function of frequency using a Gaussian process. This allows sampling at any frequency and reduces the number of required measurements. The Gaussian process is combined with a probabilistic response model to account for uncertainty in the responses. The resulting full model can be interpreted as a two-dimensional binary classifier for stimuli, and provides uncertainty bands on the estimated threshold curve. The optimal next stimulus is determined based on an information theoretic criterion. This leads to a robust adaptive estimation method that can be applied to fully automate the hearing threshold estimation process.

stat.AP↗