Bayesian optimization.

<p>(A-C) Bayesian optimization in protein sequence-function prediction. % of top-100 scores in training set found for the CNN ensemble, CNN evidential, and GP methods evaluated on the AAV/Random (A), Meltome/Random (B), and GB1/Random (C) splits. The “greedy” strategy acquired sequences with t...

وصف كامل

محفوظ في:

التفاصيل البيبلوغرافية
المؤلف الرئيسي:	Kevin P. Greenman (17703253) (author)
مؤلفون آخرون:	Ava P. Amini (15853140) (author), Kevin K. Yang (4528963) (author)
منشور في:	2025
الموضوعات:	Biochemistry Cell Biology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Chemical Sciences not elsewhere classified Information Systems not elsewhere classified select new sequences require calibrated estimations fitness landscape inference benchmarks enable us outperform greedy sampling metrics using one benchmarking uncertainty quantification retrospective active learning bayesian optimization settings bayesian optimization active learning based sampling property optimization results indicate results extend regression tasks rank correlation provide recommendations protein engineering often unable molecular machine model uncertainty model improvement learning datasets hot encoding function models effective design
الوسوم:	إضافة وسم لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!

الوصف
الملخص:	<p>(A-C) Bayesian optimization in protein sequence-function prediction. % of top-100 scores in training set found for the CNN ensemble, CNN evidential, and GP methods evaluated on the AAV/Random (A), Meltome/Random (B), and GB1/Random (C) splits. The “greedy” strategy acquired sequences with the best predicted property values. The “UCB” and “TS” strategies acquired sequences based on the upper confidence bound (UCB) and Thompson sampling (TS) approaches, respectively. The “random” strategy acquired sequences with all unseen points having equal probabilities. See the Uncertainty Methods section for an explanation of why GP experiments for the AAV landscape were not feasible. Note that in several plots, including the Gaussian process plots for Meltome and GB1 and the evidential plot for Meltome, the “greedy” strategy performance is nearly identical to and is covered up by the “UCB” strategy.</p>

Bayesian optimization.

مواد مشابهة