Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Angus Phillips

Angus Phillips

1 indexed paper

Recent (6 mo)
1
With code
0
Influential cites
0
Benchmarked
0

Publications per year

1
26

Top categories

Stats ML×1ML×1

Frequent co-authors

Tom Rossa1×
Tom Rainforth1×

Research Timeline

2026
Action-BED: Task-Driven Bayesian Experimental Design with Singly Intractable Objectives

This paper proposes a new formulation of Bayesian experimental design (BED) based on expected future loss (EFL) on downstream actions, which can be optimized using stochastic gradients and does not require explicit posterior or marginal likelihood estimation.

Highlighted terms show continued research focus across papers

Papers

stat.MLcs.LGTheoreticalRecentJun 22, 2026

Action-BED: Task-Driven Bayesian Experimental Design with Singly Intractable Objectives

Tom Rossa, Angus Phillips, Tom Rainforth

This paper proposes a new formulation of Bayesian experimental design (BED) based on expected future loss (EFL) on downstream actions, which can be optimized using stochastic gradients and does not re…

View →