Nonmyopic Multifidelity Active Search

2021/06/11 by Quan Nguyen, Nguyen, Quan, Arghavan Modiri +3 · 1 citation
Computer Science · Decision Sciences · #Advanced Bandit Algorithms Research #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning and Algorithms #Optimization and Search Problems

paper · pdf · doi:10.48550/arxiv.2106.06356

openalex publication_date 2021/06/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Active search is a learning paradigm where we seek to identify as many members of a rare, valuable class as possible given a labeling budget. Previous work on active search has assumed access to a faithful (and expensive) oracle reporting experimental results. However, some settings offer access to cheaper surrogates such as computational simulation that may aid in the search. We propose a model of multifidelity active search, as well as a novel, computationally efficient policy for this setting that is motivated by state-of-the-art classical policies. Our policy is nonmyopic and budget aware, allowing for a dynamic tradeoff between exploration and exploitation. We evaluate the performance of our solution on real-world datasets and demonstrate significantly better performance than natural benchmarks.

Citations

• A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning
• Practical Bayesian Optimization of Machine Learning Algorithms

Cited by

• Vendi Information Gain for Active Learning and its Application to Ecology

Related

• Amortized nonmyopic active search via deep imitation learning
• Nonmyopic Multiclass Active Search with Diminishing Returns for Diverse Discovery
• Efficient nonmyopic active search with applications in drug and materials discovery
• Search Improves Label for Active Learning
• Cost Aware Asynchronous Multi-Agent Active Search