Statistical Learning
MsSVT: Mixed-scale Sparse Voxel Transformer for 3D Object Detection on Point Clouds
To mitigate this gap, we present a novel Mixed-scale Sparse V oxel Transformer, named MsSVT, which can well capture both types of information simultaneously by the divide-and-conquer philosophy.
ACIL: Analytic Class-Incremental Learning with Absolute Memorization and Privacy Protection
This equality is theoretically validated.
What Makes Multi-modal Learning Better than Single (Provably) Y u Huang
The world provides us with data of multiple modalities.
Diversity Enhanced Active Learning with Strictly Proper Scoring Rules
We study acquisition functions for active learning (AL) for text classification. The Expected Loss Reduction (ELR) method focuses on a Bayesian estimate of the reduction in classification error, recently updated with Mean Objective Cost of Uncertainty (MOCU).
Neural Collapse with Normalized Features: A Geometric Analysis over the Riemannian Manifold Can Y aras
In other words, they form a Simplex Equiangular Tight Frame (ETF).
Deep Learning Through the Lens of Example Difficulty
Existing work on understanding deep learning often employs measures that compress all data-dependent information into a few numbers.
