Show newer

Learning-based Support Estimation in Sublinear Time. (arXiv:2106.08396v1 [cs.LG]) arxiv.org/abs/2106.08396

On the Sample Complexity and Metastability of Heavy-tailed Policy Search in Continuous Control. (arXiv:2106.08414v1 [cs.LG]) arxiv.org/abs/2106.08414

Online Learning with Uncertain Feedback Graphs. (arXiv:2106.08441v1 [cs.LG]) arxiv.org/abs/2106.08441

Reproducing Kernel Hilbert Space, Mercer's Theorem, Eigenfunctions, Nystr\"om Method, and Use of Kernels in Machine Learning: Tutorial and Survey. (arXiv:2106.08443v1 [stat.ML]) arxiv.org/abs/2106.08443

Gradient-trained Weights in Wide Neural Networks Align Layerwise to Error-scaled Input Correlations. (arXiv:2106.08453v1 [cs.LG]) arxiv.org/abs/2106.08453

Localized Conformal Prediction: A Generalized Inference Framework for Conformal Prediction. (arXiv:2106.08460v1 [math.ST]) arxiv.org/abs/2106.08460

Asymptotic Behavior of Common Connections in Sparse Random Networks. (arXiv:2106.08472v1 [math.ST]) arxiv.org/abs/2106.08472

A Continuized View on Nesterov Acceleration for Stochastic Gradient Descent and Randomized Gossip. (arXiv:2106.07644v1 [math.OC]) arxiv.org/abs/2106.07644

Revisiting Model Stitching to Compare Neural Representations. (arXiv:2106.07682v1 [cs.LG]) arxiv.org/abs/2106.07682

Adaptive normalization for IPW estimation. (arXiv:2106.07695v1 [stat.ME]) arxiv.org/abs/2106.07695

Robust Inference for High-Dimensional Linear Models via Residual Randomization. (arXiv:2106.07717v1 [stat.ME]) arxiv.org/abs/2106.07717

An Exponential Improvement on the Memorization Capacity of Deep Threshold Networks. (arXiv:2106.07724v1 [cs.LG]) arxiv.org/abs/2106.07724

Generalized kernel distance covariance in high dimensions: non-null CLTs and power universality. (arXiv:2106.07725v1 [math.ST]) arxiv.org/abs/2106.07725

Counterfactual Explanations as Interventions in Latent Space. (arXiv:2106.07754v1 [cs.AI]) arxiv.org/abs/2106.07754

Linear-Time Probabilistic Solutions of Boundary Value Problems. (arXiv:2106.07761v1 [stat.ML]) arxiv.org/abs/2106.07761

Improving Robustness of Graph Neural Networks with Heterophily-Inspired Designs. (arXiv:2106.07767v1 [cs.LG]) arxiv.org/abs/2106.07767

The Flip Side of the Reweighted Coin: Duality of Adaptive Dropout and Regularization. (arXiv:2106.07769v1 [cs.LG]) arxiv.org/abs/2106.07769

Bootstrapping Clustered Data in R using lmeresampler. (arXiv:2106.06568v1 [stat.ME]) arxiv.org/abs/2106.06568

Understanding Deflation Process in Over-parametrized Tensor Decomposition. (arXiv:2106.06573v1 [stat.ML]) arxiv.org/abs/2106.06573

Analysis of historical data leveraging the sandpile model of self-organized criticality demonstrates the efficacy of prescribed burns in reducing risk of destructive wildfires. (arXiv:2106.06591v1 [stat.AP]) arxiv.org/abs/2106.06591

Show older
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.