Variational Bayesian Decision-making for Continuous Utilities

Show full item record



Permalink

http://hdl.handle.net/10138/310129

Citation

Kusmierczyk , T , Sakaya , J & Klami , A 2019 , Variational Bayesian Decision-making for Continuous Utilities . in H Wallach , H Larochelle , A Beygelzimer , F d'Alché-Buc , E Fox & R Garnett (eds) , Advances in Neural Information Processing Systems 32 (NIPS 2019) . Advances in Neural Information Processing Systems , vol. 32 , Morgan Kaufmann Publishers , Maryland Heights, MO , Advances in neural information processing systems , Vancouver , Canada , 08/12/2019 . < https://papers.nips.cc/paper/8868-variational-bayesian-decision-making-for-continuous-utilities.pdf >

Title: Variational Bayesian Decision-making for Continuous Utilities
Author: Kusmierczyk, Tomasz; Sakaya, Joseph; Klami, Arto
Editor: Wallach, H.; Larochelle, H.; Beygelzimer, A.; d'Alché-Buc, F.; Fox, E.; Garnett, R.
Contributor: University of Helsinki, Department of Computer Science
University of Helsinki, Department of Computer Science
University of Helsinki, Department of Computer Science
Publisher: Morgan Kaufmann Publishers
Date: 2019-12
Language: eng
Number of pages: 11
Belongs to series: Advances in Neural Information Processing Systems 32 (NIPS 2019)
Belongs to series: Advances in Neural Information Processing Systems
URI: http://hdl.handle.net/10138/310129
Abstract: Bayesian decision theory outlines a rigorous framework for making optimal decisions based on maximizing expected utility over a model posterior. However, practitioners often do not have access to the full posterior and resort to approximate inference strategies. In such cases, taking the eventual decision-making task into account while performing the inference allows for calibrating the posterior approximation to maximize the utility. We present an automatic pipeline that co-opts continuous utilities into variational inference algorithms to account for decision-making. We provide practical strategies for approximating and maximizing the gain, and empirically demonstrate consistent improvement when calibrating approximations for specific utilities.
Subject: 113 Computer and information sciences
Rights:


Files in this item

Total number of downloads: Loading...

Files Size Format View
8868_variationa ... r_continuous_utilities.pdf 754.3Kb PDF View/Open

This item appears in the following Collection(s)

Show full item record