Knowledge-Driven Machine Learning

A Closer Look at Model Adaptation using Feature Distortion and Simplicity Bias
P. Trivedi, D. Koutra, J. J. Thiagarajan
ICLR 2023 [paper]

Single-Shot Domain Adaptation via Target-Aware Generative Augmentations
R. Subramanyam, K. Thopalli, S. Berman, P. Turaga, J. J. Thiagarajan
IEEE ICASSP 2023 [preprint] [code]

InterAug: Context-Aware Augmentations for Data-Efficient Object Detection
S. Devi, K. Thopalli, R. Dayana, P. Turaga, J. J. Thiagarajan
Under Review 2023 [preprint]

Contrastive Knowledge-Augmented Meta-Learning for Few-Shot Classification
R. Subramanyam, M. Heimann, T. S. Jayram, R.Anirudh, J. J. Thiagarajan
WACV 2023 [preprint] [code]

Learning Knowledge Graph Hierarchies for Few-Shot Dataset Generalization
R. Subramanyam, M. Heimann, T. S. Jayram, R.Anirudh, J. J. Thiagarajan
Under Review 2023 [preprint]

Analyzing Data-Centric Properties for Contrastive Learning on Graphs
P. Trivedi, M. Heimann, E. Lubana, D. Kotura, J. J. Thiagarajan
Neurips 2022 [preprint] [code]

Understanding Self-Supervised Graph Representation Learning from a Data-Centric Perspective
P. Trivedi, M. Heimann, E. Lubana, D. Kotura, J. J. Thiagarajan
KDD 2022 Workshop on Mining and Learning with Graphs [paper]

A Content-First Benchmark for Self-Supervised Graph Representation Learning
P. Trivedi, M. Heimann, E. Lubana, D. Kotura, J. J. Thiagarajan
The Webconf 2022 Workshop on Graph Learning Benchmarks [paper]

Improved StyleGAN-v2 based Inversion for Out-of-Distribution Images
R. Subramayam, V. Narayanaswamy, M. Naufel, A. Spanias, J. J. Thiagarajan
ICML 2022 [paper] [code]