Picture for Desi R. Ivanova

Desi R. Ivanova

CO-BED: Information-Theoretic Contextual Optimization via Bayesian Experimental Design

Feb 27, 2023
Desi R. Ivanova, Joel Jennings, Tom Rainforth, Cheng Zhang, Adam Foster

Figure 1 for CO-BED: Information-Theoretic Contextual Optimization via Bayesian Experimental Design

Figure 2 for CO-BED: Information-Theoretic Contextual Optimization via Bayesian Experimental Design

Figure 3 for CO-BED: Information-Theoretic Contextual Optimization via Bayesian Experimental Design

Figure 4 for CO-BED: Information-Theoretic Contextual Optimization via Bayesian Experimental Design

* 9 pages, 6 figures

Access Paper or Ask Questions

Efficient Real-world Testing of Causal Decision Making via Bayesian Experimental Design for Contextual Optimisation

Jul 12, 2022
Desi R. Ivanova, Joel Jennings, Cheng Zhang, Adam Foster

Figure 1 for Efficient Real-world Testing of Causal Decision Making via Bayesian Experimental Design for Contextual Optimisation

Figure 2 for Efficient Real-world Testing of Causal Decision Making via Bayesian Experimental Design for Contextual Optimisation

Figure 3 for Efficient Real-world Testing of Causal Decision Making via Bayesian Experimental Design for Contextual Optimisation

Figure 4 for Efficient Real-world Testing of Causal Decision Making via Bayesian Experimental Design for Contextual Optimisation

* ICML 2022 Workshop on Adaptive Experimental Design and Active Learning in the Real World. 16 pages, 5 figures

Access Paper or Ask Questions

Implicit Deep Adaptive Design: Policy-Based Experimental Design without Likelihoods

Nov 03, 2021
Desi R. Ivanova, Adam Foster, Steven Kleinegesse, Michael U. Gutmann, Tom Rainforth

Figure 1 for Implicit Deep Adaptive Design: Policy-Based Experimental Design without Likelihoods

Figure 2 for Implicit Deep Adaptive Design: Policy-Based Experimental Design without Likelihoods

Figure 3 for Implicit Deep Adaptive Design: Policy-Based Experimental Design without Likelihoods

Figure 4 for Implicit Deep Adaptive Design: Policy-Based Experimental Design without Likelihoods

* 33 pages, 8 figures. Published as a conference paper at NeurIPS 2021

Access Paper or Ask Questions