Zobrazeno 1 - 10
of 1 540
pro vyhledávání: '"Sutherland, P. J."'
Autor:
Ren, Yi, Sutherland, Danica J.
Obtaining compositional mappings is important for the model to generalize well compositionally. To better understand when and how to encourage the model to learn such mappings, we study their uniqueness through different perspectives. Specifically, w
Externí odkaz:
http://arxiv.org/abs/2409.09626
The Hilbert-Schmidt Independence Criterion (HSIC) is a powerful tool for nonparametric detection of dependence between random variables. It crucially depends, however, on the selection of reasonable kernels; commonly-used choices like the Gaussian ke
Externí odkaz:
http://arxiv.org/abs/2409.06890
We present a theoretical explanation of the ``grokking'' phenomenon, where a model generalizes long after overfitting,for the originally-studied problem of modular addition. First, we show that early in gradient descent, when the ``kernel regime'' ap
Externí odkaz:
http://arxiv.org/abs/2407.12332
The ProbCover method of Yehuda et al. is a well-motivated algorithm for active learning in low-budget regimes, which attempts to "cover" the data distribution with balls of a given radius at selected data points. We demonstrate, however, that the per
Externí odkaz:
http://arxiv.org/abs/2407.12212
Autor:
Ren, Yi, Sutherland, Danica J.
Learning dynamics, which describes how the learning of specific training examples influences the model's predictions on other examples, gives us a powerful tool for understanding the behavior of deep learning systems. We study the learning dynamics o
Externí odkaz:
http://arxiv.org/abs/2407.10490
With the widespread adoption of Large Language Models (LLMs), the prevalence of iterative interactions among these models is anticipated to increase. Notably, recent advancements in multi-round self-improving methods allow LLMs to generate new exampl
Externí odkaz:
http://arxiv.org/abs/2404.04286
We describe a data-efficient, kernel-based approach to statistical testing of conditional independence. A major challenge of conditional independence testing, absent in tests of unconditional independence, is to obtain the correct test level (the spe
Externí odkaz:
http://arxiv.org/abs/2402.13196
Autor:
Sutherland, Alexander J.
We summarize what is currently known about ed$(G)$ and RD$(G)$ for finite groups $G$ over $\mathbb{C}$ (i.e. in characteristic 0). In Appendix A, we also give an argument which improves the known bound on RD(PSL$(2,\mathbb{F}_{11})$).
Comment: 6
Comment: 6
Externí odkaz:
http://arxiv.org/abs/2312.04430
Autor:
Bae, Wonho, Ren, Yi, Ahmed, Mohamad Osama, Tung, Frederick, Sutherland, Danica J., Oliveira, Gabriel L.
Although neural networks are conventionally optimized towards zero training loss, it has been recently learned that targeting a non-zero training loss threshold, referred to as a flood level, often enables better test time generalization. Current app
Externí odkaz:
http://arxiv.org/abs/2311.02891
Most meta-learning methods assume that the (very small) context set used to establish a new task at test time is passively provided. In some settings, however, it is feasible to actively select which points to label; the potential gain from a careful
Externí odkaz:
http://arxiv.org/abs/2311.02879