Multi-distribution functionals reduce to integrals of coincidence divergences
Monotonicity under data processing and additivity on independent products force every such functional to an integral over four strata
full image
Machine Learning
Covers machine learning papers (supervised, unsupervised, semi-supervised learning, graphical models, reinforcement learning, bandits, high dimensional inference, etc.) with a statistical or theoretical grounding
Monotonicity under data processing and additivity on independent products force every such functional to an integral over four strata
full image
Contradiction Graphs Determine VC Dimension
Vertices are realizable label sequences of length m; edges mark label disagreements on shared points, fixing whether dimension meets or tops
full image
Conformal Selective Acting: Anytime-Valid Risk Control for RLVR-Trained LLMs
CSA maintains per-round selective risk bounds under predictable updates without pooling across deployments.
full image
Online Safety Monitoring for LLMs
Risk-calibrated thresholding on external verifier signals performs competitively on reasoning and red teaming tasks.
full image
The Dual Nature of LLM Persona: Aggregated Tendencies and Frame-Dependent Geometry
Aggregate trait scores resist frame changes while correlation structure drops 42% on mismatch and recovers with alignment.
full image
Cross-Audit Projection for Model Risk Prediction
Resampling audit plus asymptotic projection corrects over-optimism in binary classification risk estimates without sacrificing leading accur
full image
Aggregation with Exponential Weights is Optimal in Expectation
The bound holds for large constant temperatures on bounded Lipschitz strongly convex losses without Bernstein assumptions
full image
MLP and GNN branches stay separate until the final step, enabling direct inspection of each contribution after pretraining on larger data.
full image
Prediction Sets for Counterfactual Decisions: Coverage, Optimality, and Conformal Prediction
Equivalence to risk-averse optimization produces explicit optimal sets and a conformal method with finite-sample coverage guarantees.
full image
Conformal Bayes for Two-Sided Censored Gaussian Regression under Label Shift
Mixed atom-density calibration weights yield smaller valid sets than source-score methods in two-sided censored Gaussian regression.
full image
Sequential Structure-Sensitive Residual Diagnostics for PDE Inverse Problems
Sequential e-process rejects bad fits early using spatial residual patterns with anytime-valid error control.
full image
Born Discrete, Made Smooth: Variational Formulation of Shallow Neural Networks
A continuum variational problem on parameter densities turns training convex and yields the minimizer directly from a linear system with exp
full image
Moment-Based Selection of Multiresponse Linear Mixed-Effects Models
It reduces the problem to convex optimization using cross-moment identities and establishes finite-sample guarantees under sub-Weibull error
full image
Autorelevance function and other feature relevance measures for univariate time series
Shapley-based measures with one-step forecast replacement for missing lags identify expected patterns across ARMA and neural models.
Statistical Properties of k-means Clustering for Data Missing Completely at Random
Recovery of the true centers holds under a missing-probability and separation condition, provided centers differ in every dimension.
full image
Regularized Variational and Spectral Log-Density-Ratio Estimation in the Gaussian Location Model
In the Gaussian location model, the risk ordering reverses with the observation-to-dimension ratio under ridge regularization.
full image
Role-Aware Neural Convex Divergence Heads for Asymmetric Representation Learning
Source and target projections before convex neural divergences model asymmetric relations like entailment while keeping scores nonnegative.
full image
Distributed sources cannot be uniquely recovered from static patterns, but a transcription site allows physics-informed methods to infer the
full image
Full Bayesian Reinforcement Learning via LF-IBIS
LF-IBIS approximates posteriors over parameters and policies from simulation data alone to support uncertainty-aware decisions.
full image
Learning Effective Soliton Dynamics from Scattering Data
Weak-form identification inside the inverse scattering framework yields low-dimensional models that hold in perturbed regimes.
The benefit reverses when privacy is weaker and the usual tension with robustness returns.
full image
How to Allocate Your Tokens? Scaling Laws with Training Steps and Batch Size
Splitting training data into steps and batch size lets the scaling law fit with fewer experiments while predicting best token use.
full image
Conditional Inference Trees and Forests for Feature Selection
Benchmarks on 30 datasets show competitive top-k performance with bias-controlled split selection.
full image
From Approximation to Emergence: A Theory of Deep Learning
The account organizes results on optimization, scaling, transformers, and alignment by what each explains and what each leaves out.
full image
Decision-Aware Training for Sample-Based Generative Models
Sample-based generative models learn to penalize downstream decision costs while retaining full probabilistic output.
full image
Characterizing and Identifying Separable Graphical Models
Missing edges always admit separating sets, enabling canonical representations and an identification algorithm for equivalence classes
Function-Counting Theory for Low-Dimensional Data Structures
Extending Cover's counting theory shows how manifold structure shapes classification capacity and generalization.
full image
Deep Multitask Learning for Mixed-Type Outcomes with Shared Sparsity
Shared first layer and group Lasso yield excess risk bounds plus consistent selection of common predictors.
full image
Hierarchical Variational Kalman Filtering
Reformulated inference cuts iterations and supports higher-order trackers that become zero-phase over full history.
full image
Convolutional Symmetric AutoEncoders: enhancing latent stability via differential geometry
Extending representation consistency to convolutional layers cuts reconstruction errors and boosts robustness on advection, Burgers and Kura
full image
Approximate full-conformal multi-task regression with reproducing kernels
The construction yields a computable region guaranteed to contain the exact full-conformal one, with a volume bound when task covariances ar
full image
Active-GRPO: Adaptive Imitation and Self-Improving Reasoning for Molecular Optimization
The method switches from imitation to self-reinforcement per instance and replaces the reference with better policy candidates to exceed pri
full image
Framework tests stability of relationships when moving from structural models to machine learning adjustments on construct scores
full image
Sparse mixtures of learned prototypes keep accuracy within 2.5 points and localize influence to training neighborhoods.
full image
Ghost in the Kernel: In-Context Learning with Efficient Transformers via Domain Generalization
Domain generalization analysis shows how linear attention achieves in-context learning and informs activation choices for large-model linear
full image
Neural Network-Based Estimation of Time-Dependent Parameters in AR(p) Processes
The method keeps an explicit parametric structure while estimating changing coefficients and producing intervals under two noise distributio
full image
From Spectral Methods to Sample Complexity Bounds for Fourier Neural Operators
Bounds hold uniformly over families of equations when operators admit spectral discretizations, with rates set by smoothness and dimension.
Entropy-Regularized Probabilistic Gates for Sparse Model Discovery in Scarce-Data Federated Learning
In scarce high-dimensional data with heterogeneous clients, keeping gate entropy high yields better test accuracy and support identification
full image
Mean and variance taken from pretrained autoencoder latents produce representations that match empirical statistics more closely.
full image
The reduction yields a (1+ε) solution with Õ(min(rank(A),m)^{1/3} ε^{-2/3}) linear solves and matches ℓ_∞ regression bounds.
full image
Pivotal reduction to a scalar raises the number of tokens required, with matching bounds in each regime.
High-dimensional ODE analysis shows adversarial risk requires adaptive stepsizes unlike standard least squares.
full image
GRPO, Dr. GRPO, and DAPO Are Three Operations on One Number: The Group-Standard-Deviation Identity
Binary rewards make disagreement the exact size of each training update
full image
Uniform-in-time Propagation-of-Chaos for Stein Variational Gradient Descent
Cutoff and moment-closure arguments give log or N^{-1/2} rates that hold uniformly rather than only at short times.
Random Reshuffling Dominates Stochastic Gradient Descent
New proof removes the 1/n stepsize limit and epoch threshold that previously made theory favor standard SGD over RR.
Signed-Permutation Coordinate Transport for RMSNorm Transformers
Composing local B_d gauges along trajectories preserves SAE features and steering effects that permutation matching destroys.
full image
Accelerating Conformal Prediction via Approximate Leave-One-Out
Approximate leave-one-out estimators deliver asymptotic coverage and efficiency at far lower cost than exact refits.
STOIC reshapes STGNN residuals for zero-shot calibration and beats baselines on electricity and heating networks.
full image
Policy Optimization Achieves Data-Dependent Regret Bounds in MDPs with Unknown Transitions
The new optimistic FTRL method adds a transition complexity term but matches prior adaptive performance and recovers polylog stochastic regr
full image
On Optimal Data Splitting for Split Conformal Prediction
Analytical expressions give the training-calibration ratio that shortens intervals while keeping coverage
full image
On the Convergence of Self-Improving Online LLM Alignment
The modified objective satisfies the PL condition inside a bounded region, delivering near-linear sample complexity for LLM alignment.
full image
Contextual Slate GLM Bandits with Limited Adaptivity
Batched and rarely-switching algorithms reach O(Nd^{3/2}√T) and O(Nd√T) bounds under diversity while using only poly(N) time per round.
full image
The method bounds imputation size so that estimated clusters converge to those of the complete data.
full image
Learning Gaussian Graphical Models from a Glauber Trajectory Without Mixing
Polynomial-time recovery succeeds with trajectory length independent of mixing time under d-sparsity and minimum edge strength.
Can Tabular In-Context Learners Generalize to Biomolecular Property Prediction?
They stay competitive on ProteinGym using ESMC but performance on molecules hinges on descriptor choice.
full image
Dynamic Gaussian Processes and the Vanilla-SPDE Exchange
Vanilla-SPDE Exchange uses formulation equivalence to avoid inflating spatial dimension when observations and predictions do not coincide.
Multistage Defer Trees for Hybrid Interpretability: If at First You Can't Succeed, Tree Again
Multistage defer trees route hard cases onward while classifying the rest with one or two simple trees.
full image
Exponential-Family Tensor Completion via Nonconvex Dual Total-Variation Regularization
Upper bounds reach O(n3 rt sk^2 log / n) and close the gap to the lower bound by O(sk^2 / n) for exponential-family data.
full image
SGD at the Edge of Stability: Stochastic Stabilization with Large Learning Rates
Stochasticity forces return to stable regime in fixed steps, enabling best-iterate convergence on cross-entropy loss.
full image
Online on-policy distillation recovers only polynomial dependence and explains OPD's advantage over SFT with imperfect teachers.
full image
Dynamic Prediction of Alternating Recurrent Events via Neural Network
Inverse probability weighted pseudo-observations let the model handle censoring and dependence when predicting event-free intervals.
Stationary distribution of triplet search centers at O(ε^{-2}) with relative spread fixed by scale factor and update rule.
full image
Sequential mean-field limits turn geometric Dyson motion into a Burgers equation whose solution is the free log-normal or its convolution, m
full image
Separation Capacity of Scattering Networks
The number of binary label assignments realizable by a scattering network is controlled by its wavelet, layer, and pooling choices.
full image
Predictable GRPO: A Closed-Form Model of Training Dynamics
Mean-field reduction fixes mass, damping, and stiffness from hyperparameters plus one curvature scale, yielding group-size invariance and st
full image
Higher beta values enlarge Goodhart gaps during adaptation despite closer data support, pointing to a need for calibrated rather than maxima
full image
Optimization Dynamics Imprint Semantic Specificity in Contrastive Embedding Norms
Scale-invariant contrastive training imprints concept specificity and frequency into embedding lengths as a natural byproduct.
full image
The Fundamental Limits of Valid Transport Map Estimation
Stability assumptions make minimax lower bounds identical for all valid maps, including those produced by diffusion and flow-matching models
full image
Convergence of Continual Learning in Homogeneous Deep Networks
Weak regularization frames updates as projections onto task margins, showing local convergence is possible but global is not.
full image
ITSPACE: Monotone Gaussian Optimal Transport Updates
ITSPACE uses closed-form proximal steps on square-root factors to guarantee descent and outperform gradient methods on alignment benchmarks.
full image
Doubly Robust Adaptive Conformal Inference for Causal Effects Under Temporal Dependence
The method applies adaptive conformal inference to doubly robust pseudo-outcomes to maintain coverage when observations are serially depende
full image
Factorizable Normalizing Flows for parameter-dependent density morphing
Effects are summed at inference to handle any combination without sampling the full joint space, keeping cost linear in the number of parame
full image
Non-parametric recovery of causal diffusion mechanisms from steady-state observations
Proves non-parametric identification of the full drift function when the acyclic graph is known and supplies a consistent kernel estimator.
full image
Curvature-Weighted Gradient Diversity: A Noise Measure for Geometry-Adaptive SGD Schedules
A geometry-aware measure lets the cosine schedule adapt to directional curvature, cutting final error by up to 50 percent in diagonal quadra
full image
SGD Provably Prioritizes a Shortcut Spurious Feature in the XOR Model
Theory for two-layer ReLU networks on Boolean data shows optimization first grows the linear correlation then suppresses the quadratic signa
full image
A Stochastic--Geometric Theory of Scaling Laws in Grokking
Stopping-time analysis on the manifolds of memorization and generalization solutions predicts how learning rate, batch size and regularizati
full image
Richardson extrapolation across multiple regularized linear solves improves accuracy and stays compatible with automatic differentiation for
full image
New estimators compute the distance from projected cumulative distributions, allowing parallelism and federated aggregation without raw data
full image
When Is a Draft Accepted? A Theory of Acceptance in Speculative Decoding
Rejection regions of greedy and relaxed rules coincide with lower level sets of the target distribution, producing KL and margin bounds that
full image
Triglycerides stay unpredictable and 90% intervals under-cover some demographic groups.
full image