Search

How can we take advantage of opportunities for experimental parallelization in exploration-exploitation tradeoffs? In many experimental scenarios, it is often desirable to execute experiments simultaneously or in batches, rather than only performing one at a time. Additionally, observations may be both noisy and expensive. We introduce Gaussian Process Batch Upper Confidence Bound (GP-BUCB), an upper confidence bound-based algorithm, which models the reward function as a sample from a Gaussian process and which can select batches of experiments to run in parallel. We prove a general regret bound for GP-BUCB, as well as the surprising result that for some common kernels, the asymptotic average regret can be made independent of the batch size. The GP-BUCB algorithm is also applicable in the related case of a delay between initiation of an experiment and observation of its results, for which the same regret bounds hold. We also introduce Gaussian Process Adaptive Upper Confidence Bound (GP-AUCB), a variant of GP-BUCB which can exploit parallelism in an adaptive manner. We evaluate GP-BUCB and GP-AUCB on several simulated and real data sets. These experiments show that GP-BUCB and GP-AUCB are competitive with state-of-the-art heuristics.

Can one parallelize complex exploration exploitation tradeoffs? As an example, consider the problem of optimal high-throughput experimental design, where we wish to sequentially design batches of experiments in order to simultaneously learn a surrogate function mapping stimulus to response and identify the maximum of the function. We formalize the task as a multi-armed bandit problem, where the unknown payoff function is sampled from a Gaussian process (GP), and instead of a single arm, in each round we pull a batch of several arms in parallel. We develop GP-BUCB, a principled algorithm for choosing batches, based on the GP-UCB algorithm for sequential GP optimization. We prove a surprising result; as compared to the sequential approach, the cumulative regret of the parallel algorithm only increases by a constant factor independent of the batch size B. Our results provide rigorous theoretical support for exploiting parallelism in Bayesian global optimization. We demonstrate the effectiveness of our approach on two real-world applications., Comment: Appears in Proceedings of the 29th International Conference on Machine Learning (ICML 2012)

The COVID-19 pandemic underscored the promise of monoclonal antibody-based prophylactic and therapeutic drugs 1-3 , but also revealed how quickly viral escape can curtail effective options 4,5 . With the emergence of the SARS-CoV-2 Omicron variant in late 2021, many clinically used antibody drug products lost potency, including Evusheld ™ and its constituent, cilgavimab 4,6 . Cilgavimab, like its progenitor COV2-2130, is a class 3 antibody that is compatible with other antibodies in combination 4 and is challenging to replace with existing approaches. Rapidly modifying such high-value antibodies with a known clinical profile to restore efficacy against emerging variants is a compelling mitigation strategy. We sought to redesign COV2-2130 to rescue in vivo efficacy against Omicron BA.1 and BA.1.1 strains while maintaining efficacy against the contemporaneously dominant Delta variant. Here we show that our computationally redesigned antibody, 2130-1-0114-112, achieves this objective, simultaneously increases neutralization potency against Delta and many variants of concern that subsequently emerged, and provides protection in vivo against the strains tested, WA1/2020, BA.1.1, and BA.5. Deep mutational scanning of tens of thousands pseudovirus variants reveals 2130-1-0114-112 improves broad potency without incurring additional escape liabilities. Our results suggest that computational approaches can optimize an antibody to target multiple escape variants, while simultaneously enriching potency. Because our approach is computationally driven, not requiring experimental iterations or pre-existing binding data, it could enable rapid response strategies to address escape variants or pre-emptively mitigate escape vulnerabilities., Competing Interests: COMPETING FINANCIAL INTERESTS M.S.D. is a consultant for Inbios, Vir Biotechnology, Ocugen, Moderna and Immunome. The Diamond laboratory has received unrelated funding support in sponsored research agreements from Moderna, Vir Biotechnology, and Emergent BioSolutions. J.E.C. has served as a consultant for Luna Labs USA, Merck Sharp & Dohme Corporation, Emergent Biosolutions, and GlaxoSmithKline, is a member of the Scientific Advisory Board of Meissa Vaccines, a former member of the Scientific Advisory Board of Gigagen (Grifols) and is founder of IDBiologics. The laboratory of J.E.C. received unrelated sponsored research agreements from AstraZeneca, Takeda, and IDBiologics during the conduct of the study. J. D. B. is on the scientific advisory boards of Apriori Bio, Aerium Therapuetics, Invivyd, and the Vaccine Company. Lawrence Livermore National Laboratory, Los Alamos National Laboratory, and Vanderbilt University have applied for patents for some of the antibodies in this paper, for which T.A.D, K.T.A, A.T.Z., E.Y.L., F.Z., A.M.L., R.H.C., J.E.C., and D.M.F. are inventors. Vanderbilt University has licensed certain rights to antibodies in this paper to Astra Zeneca. J. D. B. and B.D. are inventors on Fred Hutch licensed patents related to the deep mutational scanning of viral proteins.