Knowledge-Driven Machine Learning
Jayaraman Jayaraman Thiagarajan | 21-ERD-012
Executive Summary
We will develop a novel knowledge-driven learning framework for machine learning that can leverage information from knowledge graphs without explicit supervision. The resulting advanced techniques can lead to unprecedented opportunities to improve the effectiveness of the Laboratory’s work in several strategic mission areas including biomedicine and nuclear threat reduction.
Publications, Presentations, and Patents
Subramanyam, R., M. Heimann, T. S. Jayram, R. Anirudh, and J. J. Thiagarajan. "Contrastive Knowledge-Augmented Meta-Learning for Few-Shot Classification." WACV 2023. Waikolao, HI. Accepted. 2023. LLNL-CONF-833730.
Trivedi, P., E. Lubana, M. Heimann, D. Koutra, and J. J. Thiagarajan. "Understanding Self-Supervised Graph Representation Learning from a Data-Centric Perspective." NeurIPS 2022. New Orleans, LA. Accepted. 2022. LLNL-CONF-835530.
Trivedi, P., D. Koutra, J. J. Thiagarajan. "Exploring the Design of Adaptation Protocols for Improved Generalization and Machine Learning Safety." ICML PODS 2022. Baltimore, MD. July 2022. LLNL-PROC-836992.
Subramanyam, R., V. Narayanaswamy, M. Naufel, A. Spanias, and J. J. Thiagarajan. "Improved StyleGAN-v2 Based Inversion for Out-of-Distribution Images." ICML 2022. Baltimore, MD. July 2022. LLNL-CONF-829448.
Thopalli, K., P. Turaga, and J. J. Thiagarajan. "Re-labeling Domains Improves Multi-Domain Generalization." IEEE ICASSP 2022. Singapore, May 2022. LLNL-CONF-827848.
