MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202641123325 A) filed by Mrs. Boddu Revathi; Deepika; Gargi Mishra; Dr. G. Venkatesh; Dr. G. Sathya Priyanka; and Dr. T. O. Sunitha on October 14, 2026, for Method And System For Bayesian Optimization Of Ai Models In Drug Discovery.
Inventors include Mrs. Boddu Revathi; Deepika; Gargi Mishra; Dr. G. Venkatesh; Dr. G. Sathya Priyanka; and Dr. T. O. Sunitha.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: A system and method for Bayesian optimization of artificial-intelligence-based drug discovery predictions comprises a molecular data storage unit, molecular structure processing unit, training data processing unit, probabilistic prediction processor, acquisition processing unit, candidate validation unit, evaluation interface unit, feedback processing unit and control processor. The system processes molecular structures and associated evaluation data to generate structural feature representations, predicted molecular property values and corresponding prediction uncertainties. The acquisition processing unit determines acquisition values using the predicted property values and uncertainties and selects candidate molecular structures for computational or experimental evaluation. The candidate validation unit verifies structural and physicochemical constraints before evaluation. Received evaluation results are associated with corresponding molecular structures and incorporated into updated evaluated molecular records. The probabilistic prediction processor subsequently generates revised predictions and uncertainties, while the control processor repeatedly performs candidate selection and evaluation subject to evaluation budget and convergence conditions.
Disclaimer: Curated by HT Syndication.
