Bounded overestimation of the number of factors in PCA preserves √T-valid inference and consistent factor-space recovery under a random-matrix local law.
hub
Advances in neural information processing systems , volume=
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
years
2026 12roles
background 1polarities
background 1representative citing papers
A solvable hierarchical model with power-law feature strengths yields explicit power-law scaling of prediction error through sequential recovery of latent directions by a layer-wise spectral algorithm.
Sobolev regularization on the witness function enables global convergence of MMD gradient flows for both sampling and generative modeling without isoperimetric assumptions.
ALiBi bias is the expectation of positional LSH-induced block masks, yielding spectral and max-norm approximation bounds that reduce long-context biased attention to randomized short-context unbiased attention.
New query-time bound of tilde O(d + epsilon Delta squared + 1/epsilon cubed) for Gaussian kernel mean estimation, improving prior bounds for small epsilon and intermediate diameter via a fast spherical embedding theorem.
CCBO enables collaborative contextual Bayesian optimization across clients with sublinear regret guarantees and shows substantial gains over non-collaborative methods in simulations and a hot rolling application even under heterogeneity.
AdamO modifies Adam with an orthogonality correction to ensure the spectral radius of the TD update operator stays below one, providing a theoretical stability guarantee for offline RL.
An actor-critic RL algorithm for low-rank MDPs achieves improved sample efficiency using solely a policy evaluation oracle.
Proposes an e-process-based sequential diagnostic that detects misspecified PDE inverse problem fits earlier than standard discrepancy methods while providing anytime-valid type-I error control.
PEQ-Net uses policy-aware reparameterization of ICE Q-functions and kernel mean embeddings in a shared encoder, followed by LTMLE, to jointly estimate multiple policies while constraining second-order bias for lower variance.
Quantum feature surrogates let a quantum processor act as a teacher on a small data subsample while a classical surrogate applies the learned representations to the entire industrial dataset.
A mechanics of the learning process is emerging in deep learning theory, characterized by dynamics, coarse statistics, and falsifiable predictions across idealized settings, limits, laws, hyperparameters, and universal behaviors.
citing papers explorer
-
Fixed-order PCA: Theory for Overestimated Factor Models
Bounded overestimation of the number of factors in PCA preserves √T-valid inference and consistent factor-space recovery under a random-matrix local law.
-
Scaling Laws from Sequential Feature Recovery: A Solvable Hierarchical Model
A solvable hierarchical model with power-law feature strengths yields explicit power-law scaling of prediction error through sequential recovery of latent directions by a layer-wise spectral algorithm.
-
Sobolev Regularized MMD Gradient Flow
Sobolev regularization on the witness function enables global convergence of MMD gradient flows for both sampling and generative modeling without isoperimetric assumptions.
-
Positional LSH: Binary Block Matrix Approximation for Attention with Linear Biases
ALiBi bias is the expectation of positional LSH-induced block masks, yielding spectral and max-norm approximation bounds that reduce long-context biased attention to randomized short-context unbiased attention.
-
New Bounds for Kernel Sums via Fast Spherical Embeddings
New query-time bound of tilde O(d + epsilon Delta squared + 1/epsilon cubed) for Gaussian kernel mean estimation, improving prior bounds for small epsilon and intermediate diameter via a fast spherical embedding theorem.
-
Collaborative Contextual Bayesian Optimization
CCBO enables collaborative contextual Bayesian optimization across clients with sublinear regret guarantees and shows substantial gains over non-collaborative methods in simulations and a hot rolling application even under heterogeneity.
-
AdamO: A Collapse-Suppressed Optimizer for Offline RL
AdamO modifies Adam with an orthogonality correction to ensure the spectral radius of the TD update operator stays below one, providing a theoretical stability guarantee for offline RL.
-
Breaking the Computational Barrier: Provably Efficient Actor-Critic for Low-Rank MDPs
An actor-critic RL algorithm for low-rank MDPs achieves improved sample efficiency using solely a policy evaluation oracle.
-
Sequential Structure-Sensitive Residual Diagnostics for PDE Inverse Problems
Proposes an e-process-based sequential diagnostic that detects misspecified PDE inverse problem fits earlier than standard discrepancy methods while providing anytime-valid type-I error control.
-
Smooth Multi-Policy Causal Effect Estimation in Longitudinal Settings
PEQ-Net uses policy-aware reparameterization of ICE Q-functions and kernel mean embeddings in a shared encoder, followed by LTMLE, to jointly estimate multiple policies while constraining second-order bias for lower variance.
-
Off-line quantum-advantage feature extraction for industrial production
Quantum feature surrogates let a quantum processor act as a teacher on a small data subsample while a classical surrogate applies the learned representations to the entire industrial dataset.
-
There Will Be a Scientific Theory of Deep Learning
A mechanics of the learning process is emerging in deep learning theory, characterized by dynamics, coarse statistics, and falsifiable predictions across idealized settings, limits, laws, hyperparameters, and universal behaviors.