Research Areas
My research focuses on developing efficient machine learning algorithms for computer vision and natural language processing.
Few-Shot Learning with Meta-Learning
Developing algorithms that can learn from very few examples by leveraging meta- learning approaches.
Impact:
Reduced the number of required training examples by 80% while maintaining model performance.
Self-Supervised Representation Learning
Creating methods for learning useful representations without labeled data through self-supervision.
Impact:
Achieved state-of-the-art results on benchmark datasets with 40% less labeled data.
Efficient Deep Learning
Designing lightweight neural network architectures that maintain performance while reducing computational costs.
Impact:
Reduced inference time by 65% and model size by 70% with minimal accuracy loss.