Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 2 of 2 for “"Dataset Distillation"”.
-
TOWARDS DATA-EFFICIENT DEEP LEARNING
… learning by tackling the limitations of current dataset distillation (DD) methods, which aim to compress large datasets into compact synthetic ones for faster training and enhanced privacy. First, it introduces Dataset Factorization, a novel framework that decomposes datasets into learnable bases …
-
Designing Provably Convergent Algorithms from the Geometry of Data
… is then used to derive a provably convergent dataset distillation technique, which finds a small training set that maximises performance after training. These algorithms are derived using techniques from Monte Carlo methods, Wasserstein metrics, optimal quantisation, and diffusion processes.