Bayesian optimization and attribute adjustment

Stephan Eismann, Daniel Levy, Rui Shu, Stefan Bartzsch, Stefano Ermon
Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence, PMLR R16:1041-1051, 2018.

Abstract

Automatic design via Bayesian optimization holds great promise given the constant increase of available data across domains. However, it faces difficulties from high-dimensional, poten- tially discrete, search spaces. We propose to probabilistically embed inputs into a lower di- mensional, continuous latent space, where we perform gradient-based optimization guided by a Gaussian process. Building on variational au- toncoders, we use both labeled and unlabeled data to guide the encoding and increase its ac- curacy. In addition, we propose an adversar- ial extension to render the latent representa- tion invariant with respect to specific design attributes, which allows us to transfer these at- tributes across structures. We apply the frame- work both to a functional-protein dataset and to perform optimization of drag coefficients di- rectly over high-dimensional shapes without in- corporating domain knowledge or handcrafted features.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR16-eismann18a, title = {{B}ayesian optimization and attribute adjustment}, author = {Eismann, Stephan and Levy, Daniel and Shu, Rui and Bartzsch, Stefan and Ermon, Stefano}, booktitle = {Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence}, pages = {1041--1051}, year = {2018}, editor = {Globerson, Amir and Silva, Ricardo}, volume = {R16}, series = {Proceedings of Machine Learning Research}, month = {06--10 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r16/main/assets/eismann18a/eismann18a.pdf}, url = {https://proceedings.mlr.press/r16/eismann18a.html}, abstract = {Automatic design via Bayesian optimization holds great promise given the constant increase of available data across domains. However, it faces difficulties from high-dimensional, poten- tially discrete, search spaces. We propose to probabilistically embed inputs into a lower di- mensional, continuous latent space, where we perform gradient-based optimization guided by a Gaussian process. Building on variational au- toncoders, we use both labeled and unlabeled data to guide the encoding and increase its ac- curacy. In addition, we propose an adversar- ial extension to render the latent representa- tion invariant with respect to specific design attributes, which allows us to transfer these at- tributes across structures. We apply the frame- work both to a functional-protein dataset and to perform optimization of drag coefficients di- rectly over high-dimensional shapes without in- corporating domain knowledge or handcrafted features.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Bayesian optimization and attribute adjustment %A Stephan Eismann %A Daniel Levy %A Rui Shu %A Stefan Bartzsch %A Stefano Ermon %B Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2018 %E Amir Globerson %E Ricardo Silva %F pmlr-vR16-eismann18a %I PMLR %P 1041--1051 %U https://proceedings.mlr.press/r16/eismann18a.html %V R16 %X Automatic design via Bayesian optimization holds great promise given the constant increase of available data across domains. However, it faces difficulties from high-dimensional, poten- tially discrete, search spaces. We propose to probabilistically embed inputs into a lower di- mensional, continuous latent space, where we perform gradient-based optimization guided by a Gaussian process. Building on variational au- toncoders, we use both labeled and unlabeled data to guide the encoding and increase its ac- curacy. In addition, we propose an adversar- ial extension to render the latent representa- tion invariant with respect to specific design attributes, which allows us to transfer these at- tributes across structures. We apply the frame- work both to a functional-protein dataset and to perform optimization of drag coefficients di- rectly over high-dimensional shapes without in- corporating domain knowledge or handcrafted features. %Z Reissued by PMLR on 04 October 2026.
APA
Eismann, S., Levy, D., Shu, R., Bartzsch, S. & Ermon, S.. (2018). Bayesian optimization and attribute adjustment. Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R16:1041-1051 Available from https://proceedings.mlr.press/r16/eismann18a.html. Reissued by PMLR on 04 October 2026.

Related Material