Stefano Soatto

Incremental Learning for Metric-Based Meta-Learners

2020 · arXiv (Cornell University)

Majority of the modern meta-learning methods for few-shot classification tasks operate in two phases: a meta-training phase where the meta-learner learns a generic representation by solving multiple few-shot tasks sampled from a large dataset and …

Tangent Model Composition for Ensembling and Continual Fine-tuning

2023 · arXiv (Cornell University)

Tangent Model Composition (TMC) is a method to combine component models independently fine-tuned around a pre-trained point. Component models are tangent vectors to the pre-trained model that can be added, scaled, or subtracted to support …