Improving forward compatibility in class incremental learning by increasing representation rank and feature richness.

Class Incremental Learning (CIL) constitutes a pivotal subfield within continual learning, aimed at enabling models to progressively learn new classification tasks while retaining knowledge obtained from prior tasks. Although previous studies have predominantly focused on backward compatible approaches to mitigate catastrophic forgetting, recent investigations have introduced forward compatible methods to enhance performance on novel tasks and complement existing backward compatible methods. In this study, we introduce effective-Rank based Feature Richness enhancement (RFR) method that is designed for improving forward compatibility.

An effective low-rank compression with a joint rank selection followed by a compression-friendly training.

Low-rank compression of a neural network is one of the popular compression techniques, where it has been known to have two main challenges. The first challenge is determining the optimal rank of all the layers and the second is training the neural network into a compression-friendly form. To overcome the two challenges, we propose BSR (Beam-search and Stable Rank), a low-rank compression algorithm that embodies an efficient rank-selection method and a unique compression-friendly training method.

Quantification of electron transfer rates of different facets on single gold nanoparticles during catalytic reactions.

Rayleigh scattering spectra of high-index {730} elongated tetrahexahedral gold nanoparticles and low-index {100}, {110}, and {111} gold nanorods were collected in real time in the reduction of 4-nitrophenol. The high-index facets are capable of accepting electrons seven times faster and emitting electrons two-and-a-half times faster than low-index facets.