MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641085237 A) filed by Cmr Technical Campus; Cmr College Of Engineering & Technology; and Cmr Institute Of Technology on July 11, 2026, for Self-Reflective Neural Network With Internal Counterfactual Generation And Gradient Reweighting For Bias-Aware Adaptive Learning.
Inventors include S. Ram Chandra Reddy; B. Venkatesh; Ms. B. Sujani; Ms. D. Madhavi; Mrs. V. Siva Pavani; and Mr. D. Jyothi.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: A system (100) and method for bias-aware adaptive learning using a self-reflective neural network are disclosed. The system (100) comprises a memory (102), a processor (104), and a communication module (106). The memory (102) stores instructions executable by the processor (104) to implement a data acquisition module (108), a counterfactual generation module (110), a self-reflection module (112), and a gradient reweighting module (114). The data acquisition module (108) receives training and operational data for machine learning tasks. The counterfactual generation module (110) internally generates alternative representations corresponding to predictive outputs. The self-reflection module (112) evaluates learned feature representations to identify bias-associated learning characteristics. The gradient reweighting module (114) dynamically adjusts optimization gradients to mitigate identified bias influences. The processor (104) adapts neural network parameters using reweighted gradients to improve fairness, robustness, adaptive learning capability, and prediction reliability across machine learning environments.
Disclaimer: Curated by HT Syndication.
