Combining Latent Space and Structured Kernels for Bayesian Optimization over Combinatorial Spaces
Dec 6, 2021
Speakers
Aryan Deshwal
Speaker · 0 followers
Jana Doppa
Speaker · 0 followers
About
We consider the problem of optimizing combinatorial spaces (e.g., sequences, trees, and graphs) using expensive black-box function evaluations. For example, optimizing molecules for drug design using physical lab experiments. Bayesian optimization (BO) is an efficient framework for solving such problems by intelligently selecting the inputs with high utility guided by a learned surrogate model. A recent BO approach for combinatorial spaces is through a reduction to BO over continuous spaces via learning a latent representation from a database of unsupervised structures. The selected input from the continuous space is decoded into a discrete structure for performing function evaluation. However, this BO procedure doesn't take into account the corresponding decoded structures of points in the latent space resulting in a disconnect between the actual function evaluation and the BO process. To overcome this drawback, this paper proposes a principled approach referred as LADDER (Latent Space BO guided by Decoded Structures). The key idea is to integrate the structural information from decoded structures with the learned latent space representation for better surrogate modeling. Our experiments on real-world benchmarks show that LADDER significantly improves over baseline methods.We consider the problem of optimizing combinatorial spaces (e.g., sequences, trees, and graphs) using expensive black-box function evaluations. For example, optimizing molecules for drug design using physical lab experiments. Bayesian optimization (BO) is an efficient framework for solving such problems by intelligently selecting the inputs with high utility guided by a learned surrogate model. A recent BO approach for combinatorial spaces is through a reduction to BO over continuous spaces via…
Organizer
NeurIPS 2021
Account · 1.9k followers
About NeurIPS 2021
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Opening Remarks
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Consistent Non-Parametric Methods for Maximizing Robustness
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Model Adaptation: Historical Contrastive Learning for Unsupervised Domain Adaptation without Source Data
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Geometry Processing with Neural Fields
Watch later
Guandao Yang, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Casual discovery by generative modelling
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
How Should Pre-Trained Language Models Be Fine-Tuned Towards Adversarial Robustness?
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%