AISTATS 2025 Schedule

Filter Events

FRI 2 MAY

6 p.m.

Registration Desk

(ends 2:00 AM)

6:30 p.m.

Opening Remarks:

Opening Remarks

(ends 7:00 PM)

7 p.m.

Invited Talk:

Keynote: Chris Holmes

Chris Holmes

(ends 8:00 PM)

8 p.m.

Break:

Coffee Break

(ends 8:30 PM)

8:30 p.m.

Oral Session 1: Deep Learning and Learning Theory [8:30-9:30]

Orals -

A Random Matrix Theory Perspective on the Spectrum of Learned Features and Asymptotic Generalization Capabilities

Learning Graph Node Embeddings by Smooth Pair Sampling

Loss Gradient Gaussian Width based Generalization and Optimization Guarantees

Symmetry-Based Structured Matrices for Efficient Approximately Equivariant Networks

What Ails Generative Structure-based Drug Design: Expressivity is Too Little or Too Much?

(ends 9:30 PM)

9:30 p.m.

Panel:

Discussion Panel

(ends 10:30 PM)

10:30 p.m.

Break:

Lunch + DEI Events

(ends 12:00 AM)

SAT 3 MAY

midnight

Oral Session 2: Distribution Learning and Causality [12:00-1:00]

Orals -

Additive Model Boosting: New Insights and Path(ologie)s

Causal discovery in mixed additive noise models

Distributional Counterfactual Explanations With Optimal Transport

Importance-weighted Positive-unlabeled Learning for Distribution Shift Adaptation

On Distributional Discrepancy for Experimental Design with General Assignment Probabilities

(ends 1:00 AM)

1 a.m.

Poster Session 1 [1:00-4:00]

Posters -

Causal discovery in mixed additive noise models

Almost linear time differentially private release of synthetic graphs

InfoNCE: Identifying the Gap Between Theory and Practice

Differentially private algorithms for linear queries via stochastic convex optimization

Infinite-dimensional Diffusion Bridge Simulation via Operator Learning

Robust Estimation in metric spaces: Achieving Exponential Concentration with a Fr\'echet Median

Fine-Tuning with Uncertainty-Aware Priors Makes Vision and Language Foundation Models More Reliable

Gaussian Smoothing in Saliency Maps: The Stability-Fidelity Trade-Off in Neural Network Interpretability

Multi-Player Approaches for Dueling Bandits

Distribution-Aware Mean Estimation under User-level Local Differential Privacy

A Tight Regret Analysis of Non-Parametric Repeated Contextual Brokerage

On Local Posterior Structure in Deep Ensembles

Multi-marginal Schrödinger Bridges with Iterative Reference Refinement

Common Learning Constraints Alter Interpretations of Direct Preference Optimization

A High Dimensional Statistical Model for Adversarial Training: Geometry and Trade-Offs

Fair Resource Allocation in Weakly Coupled Markov Decision Processes

Fast Convergence of Softmax Policy Mirror Ascent

Deep Generative Quantile Bayes

Effective Bayesian Causal Inference via Structural Marginalisation and Autoregressive Orders

Hierarchical Bias-Driven Stratification for Interpretable Causal Effect Estimation

Learning a Single Index Model from Anisotropic Data with Vanilla Stochastic Gradient Descent

All or None: Identifiable Linear Properties of Next-Token Predictors in Language Modeling

Pareto Set Identification With Posterior Sampling

Variational Inference in Location-Scale Families: Exact Recovery of the Mean and Correlation Matrix

Looped ReLU MLPs May Be All You Need as Practical Programmable Computers

On the Asymptotic Mean Square Error Optimality of Diffusion Models

Steinmetz Neural Networks for Complex-Valued Data

Theoretical Analysis of Leave-one-out Cross Validation for Non-differentiable Penalties under High-dimensional Settings

Optimal downsampling for Imbalanced Classification with Generalized Linear Models

An Empirical Bernstein Inequality for Dependent Data in Hilbert Spaces and Applications

Training LLMs with MXFP4

Time-series attribution maps with regularized contrastive learning

Statistical Inference for Feature Selection after Optimal Transport-based Domain Adaptation

Epistemic Uncertainty and Excess Risk in Variational Inference

Distributional Counterfactual Explanations With Optimal Transport

Independent Learning in Performative Markov Potential Games

Prior-Dependent Allocations for Bayesian Fixed-Budget Best-Arm Identification in Structured Bandits

Fundamental Limits of Perfect Concept Erasure

Bandit Pareto Set Identification in a Multi-Output Linear Model

Analyzing the Role of Permutation Invariance in Linear Mode Connectivity

Sketch-and-Project Meets Newton Method: Global O(1/k^2) Convergence with Low-Rank Updates

DPFL: Decentralized Personalized Federated Learning

Efficient Estimation of a Gaussian Mean with Local Differential Privacy

On Preference-based Stochastic Linear Contextual Bandits with Knapsacks

How Well Can Transformers Emulate In-Context Newton's Method?

Hybrid Transfer Reinforcement Learning: Provable Sample Efficiency from Shifted-Dynamics Data

Approximating the Total Variation Distance between Gaussians

Incremental Uncertainty-aware Performance Monitoring with Active Labeling Intervention

Wasserstein Gradient Flow over Variational Parameter Space for Variational Inference

ChronosX: Adapting Pretrained Time Series Models with Exogenous Variables

Towards Regulatory-Confirmed Adaptive Clinical Trials: Machine Learning Opportunities and Solutions

Application of Structured State Space Models to High energy physics with locality sensitive hashing

Optimizing Neural Network Training and Quantization with Rooted Logistic Objectives

Linear Submodular Maximization with Bandit Feedback

Fixed-Budget Change Point Identification in Piecewise Constant Bandits

LMEraser: Large Model Unlearning via Adaptive Prompt Tuning

Symmetry-Based Structured Matrices for Efficient Approximately Equivariant Networks

Randomized Iterative Solver as Iterative Refinement: A Simple Fix Towards Backward Stability

Memorization in Attention-only Transformers

Keeping up with dynamic attackers: Certifying robustness to adaptive online data poisoning

Scalable spectral representations for multiagent reinforcement learning in network MDPs

Bayes without Underfitting: Fully Correlated Deep Learning Posteriors via Alternating Projections

Out-of-distribution robustness for multivariate analysis via causal regularisation

Changepoint Estimation in Sparse Dynamic Stochastic Block Models under Near-Optimal Signal Strength

Federated Communication-Efficient Multi-Objective Optimization

Deep Clustering via Probabilistic Ratio-Cut Optimization

Recurrent Neural Goodness-of-Fit Test for Time Series

Statistical Learning of Distributionally Robust Stochastic Control in Continuous State Spaces

A Causal Framework for Evaluating Deferring Systems

A Bias-Variance Decomposition for Ensembles over Multiple Synthetic Datasets

A Shared Low-Rank Adaptation Approach to Personalized RLHF

Transformers are Provably Optimal In-context Estimators for Wireless Communications

Bayesian Principles Improve Prompt Learning In Vision-Language Models

Sampling in High-Dimensions using Stochastic Interpolants and Forward-Backward Stochastic Differential Equations

Minimum Empirical Divergence for Sub-Gaussian Linear Bandits

Signature Isolation Forest

Reinforcement Learning with Intrinsically Motivated Feedback Graph for Lost-sales Inventory Control

Class Imbalance in Anomaly Detection: Learning from an Exactly Solvable Model

Offline Multi-task Transfer RL with Representational Penalization

Counting Graphlets of Size k under Local Differential Privacy

Learning the Distribution Map in Reverse Causal Performative Prediction

Risk-sensitive Bandits: Arm Mixture Optimality and Regret-efficient Algorithms

On the Convergence of Locally Adaptive and Scalable Diffusion-Based Sampling Methods for Deep Bayesian Neural Network Posteriors

SubSearch: Robust Estimation and Outlier Detection for Stochastic Block Models via Subgraph Search

FLIPHAT: Joint Differential Privacy for High Dimensional Linear Bandits

General Staircase Mechanisms for Optimal Differential Privacy

Legitimate ground-truth-free metrics for deep uncertainty classification scoring

TempTest: Local Normalization Distortion and the Detection of Machine-generated Text

Optimal Stochastic Trace Estimation in Generative Modeling

Sample Compression Unleashed: New Generalization Bounds for Real Valued Losses

Unifying Feature-Based Explanations with Functional ANOVA and Cooperative Game Theory

Adaptive RKHS Fourier Features for Compositional Gaussian Process Models

Learning Infinite-Horizon Average-Reward Linear Mixture MDPs of Bounded Span

A Multi-Task Learning Approach to Linear Multivariate Forecasting

Model selection for behavioral learning data and applications to contextual bandits

Microfoundation inference for strategic prediction

Cross-modality Matching and Prediction of Perturbation Responses with Labeled Gromov-Wasserstein Optimal Transport

Bayesian Gaussian Process ODEs via Double Normalizing Flows

Statistical Guarantees for Lifelong Reinforcement Learning using PAC-Bayes Theory

Towards Fair Graph Learning without Demographic Information

Estimation of Large Zipfian Distributions with Sort and Snap

On the Computational Tractability of the (Many) Shapley Values

MDP Geometry, Normalization and Reward Balancing Solvers

Prior-Fitted Networks Scale to Larger Datasets When Treated as Weak Learners

Improved dependence on coherence in eigenvector and eigenvalue estimation error bounds

Automatically Adaptive Conformal Risk Control

Variational Adversarial Training Towards Policies with Improved Robustness

Near-Optimal Algorithm for Non-Stationary Kernelized Bandits

Bridging Domains with Approximately Shared Features

Gaussian Mean Testing under Truncation

Global Ground Metric Learning with Applications to scRNA data

Amortized Probabilistic Conditioning for Optimization, Simulation and Inference

Bypassing the Exponential Dependency: Looped Transformers Efficiently Learn In-context by Multi-step Gradient Descent

Locally Private Estimation with Public Features

Local Stochastic Sensitivity Analysis For Dynamical Systems

Harnessing the Power of Vicinity-Informed Analysis for Classification under Covariate Shift

Planning and Learning in Risk-Aware Restless Multi-Arm Bandits

Relating Piecewise Linear Kolmogorov Arnold Networks to ReLU Networks

A Random Matrix Theory Perspective on the Spectrum of Learned Features and Asymptotic Generalization Capabilities

Exposing Privacy Gaps: Membership Inference Attack on Preference Data for LLM Alignment

Cross-Modal Imputation and Uncertainty Estimation for Spatial Transcriptomics

Evaluating Prediction-based Interventions with Human Decision Makers In Mind

Distributional Adversarial Loss

A Generalized Theory of Mixup for Structure-Preserving Synthetic Data

Conditional Prediction ROC Bands for Graph Classification

Steering No-Regret Agents in MFGs under Model Uncertainty

Quantifying the Optimization and Generalization Advantages of Graph Neural Networks Over Multilayer Perceptrons

Characterizing the Accuracy-Communication-Privacy Trade-off in Distributed Stochastic Convex Optimization

Narrowing the Gap between Adversarial and Stochastic MDPs via Policy Optimization

Learning Graph Node Embeddings by Smooth Pair Sampling

Information-Theoretic Causal Discovery in Topological Order

Bridging the Theoretical Gap in Randomized Smoothing

Unbiased Quantization of the $L_1$ Ball for Communication-Efficient Distributed Mean Estimation

Differentiable Causal Structure Learning with Identifiability by NOTIME

Differential Privacy in Distributed Learning: Beyond Uniformly Bounded Stochastic Gradients

Tighter Confidence Bounds for Sequential Kernel Regression

Proximal Sampler with Adaptive Step Size

Learning to Negotiate via Voluntary Commitment

Conditional Generative Learning from Invariant Representations in Multi-Source: Robustness and Efficiency

Hypernym Bias: Unraveling Deep Classifier Training Dynamics through the Lens of Class Hierarchy

The Sample Complexity of Stackelberg Games

Differentially Private Continual Release of Histograms and Related Queries

Type Information-Assisted Self-Supervised Knowledge Graph Denoising

Convergence Analysis for General Probability Flow ODEs of Diffusion Models in Wasserstein Distances

Behavior-Inspired Neural Networks for Relational Inference

The Size of Teachers as a Measure of Data Complexity: PAC-Bayes Excess Risk Bounds and Scaling Laws

Robust Score Matching

Signal Recovery from Random Dot-Product Graphs under Local Differential Privacy

MEDUSA: Medical Data Under Shadow Attacks via Hybrid Model Inversion

Bayesian Off-Policy Evaluation and Learning for Large Action Spaces

Efficient Exploitation of Hierarchical Structure in Sparse Reward Reinforcement Learning

Importance-weighted Positive-unlabeled Learning for Distribution Shift Adaptation

Locally Private Sampling with Public Data

A Safe Bayesian Learning Algorithm for Constrained MDPs with Bounded Constraint Violation

Near-optimal algorithms for private estimation and sequential testing of collision probability

Order-Optimal Regret in Distributed Kernel Bandits

QuACK: A Multipurpose Queuing Algorithm for Cooperative $k$-Armed Bandits

Variance-Dependent Regret Bounds for Nonstationary Linear Bandits

Cost-aware simulation-based inference

Optimal Multi-Objective Best Arm Identification with Fixed Confidence

Time-varying Gaussian Process Bandits with Unknown Prior

Online Assortment and Price Optimization Under Contextual Choice Models

Variation Due to Regularization Tractably Recovers Bayesian Deep Learning Uncertainty

An Adaptive Method for Weak Supervision with Drifting Data

Provable Benefits of Task-Specific Prompts for In-context Learning

Spectral Representation for Causal Estimation with Hidden Confounders

Faster WIND: Accelerating Iterative Best-of-$N$ Distillation for LLM Alignment

Privacy in Metalearning and Multitask Learning: Modeling and Separations

Invariant Link Selector for Spatial-Temporal Out-of-Distribution Problem

Learning Stochastic Nonlinear Dynamics with Embedded Latent Transfer Operators

Reinforcement Learning for Adaptive MCMC

Level Set Teleportation: An Optimization Perspective

Choice is what matters after Attention

Unbiased and Sign Compression in Distributed Learning: Comparing Noise Resilience via SDEs

Online-to-PAC generalization bounds under graph-mixing dependencies

Functional Stochastic Gradient MCMC for Bayesian Neural Networks

Fundamental computational limits of weak learnability in high-dimensional multi-index models

Batch, match, and patch: low-rank approximations for score-based variational inference

Recursive Learning of Asymptotic Variational Objectives

Knowledge Graph Completion with Mixed Geometry Tensor Factorization

SteinDreamer: Variance Reduction for Text-to-3D Score Distillation via Stein Identity

AlleNoise - large-scale text classification benchmark dataset with real-world label noise

Is Prior-Free Black-Box Non-Stationary Reinforcement Learning Feasible?

A Theoretical Understanding of Chain-of-Thought: Coherent Reasoning and Error-Aware Demonstration

Balls-and-Bins Sampling for DP-SGD

The Polynomial Iteration Complexity for Variance Exploding Diffusion Models: Elucidating SDE and ODE Samplers

Information-Theoretic Measures on Lattices for Higher-Order Interactions

Function-Space MCMC for Bayesian Wide Neural Networks

Stochastic Approximation with Unbounded Markovian Noise: A General-Purpose Theorem

Towards Cost Sensitive Decision Making

Emergence of Globally Attracting Fixed Points in Deep Neural Networks With Nonlinear Activations

Your Finetuned Large Language Model is Already a Powerful Out-of-distribution Detector

HACSurv: A Hierarchical Copula-Based Approach for Survival Analysis with Dependent Competing Risks

Fourier Circuits in Neural Networks and Transformers: A Case Study of Modular Arithmetic with Multiple Inputs

HAR-former: Hybrid Transformer with an Adaptive Time-Frequency Representation Matrix for Long-Term Series Forecasting

(ends 4:00 AM)

4 a.m.

Reception:

Opening Reception

(ends 6:00 AM)

6:30 p.m.

Registration Desk

(ends 2:00 AM)

7 p.m.

Invited Talk:

Keynote: Akshay Krishnamurthy

Akshay Krishnamurthy

(ends 8:00 PM)

8 p.m.

Break:

Coffee Break

(ends 8:30 PM)

8:30 p.m.

Oral Session 3: Optimization [8:30-9:30]

Orals -

Cubic regularized subspace Newton for non-convex optimization

Implicit Diffusion: Efficient optimization through stochastic sampling

ScoreFusion: Fusing Score-based Generative Models via Kullback–Leibler Barycenters

The Pivoting Framework: Frank-Wolfe Algorithms with Active Set Size Control

Unbiased and Sign Compression in Distributed Learning: Comparing Noise Resilience via SDEs

(ends 9:30 PM)

9:30 p.m.

Awards:

Three Awards Talks

(ends 10:30 PM)

10:30 p.m.

Break:

Lunch + DEI Events

(ends 12:00 AM)

SUN 4 MAY

midnight

Oral Session 4: Privacy and Games [12:00-1:00]

Orals -

Almost linear time differentially private release of synthetic graphs

A Novel Convex Gaussian Min Max Theorem for Repeated Features

Balls-and-Bins Sampling for DP-SGD

Some Targets Are Harder to Identify than Others: Quantifying the Target-dependent Membership Leakage

The Sample Complexity of Stackelberg Games

(ends 1:00 AM)

1 a.m.

Poster Session 2 [1:00-4:00]

Posters -

Some Targets Are Harder to Identify than Others: Quantifying the Target-dependent Membership Leakage

SNAP: Sequential Non-Ancestor Pruning for Targeted Causal Effect Estimation With an Unknown Graph

Flexible Copula-Based Mixed Models in Deep Learning: A Scalable Approach to Arbitrary Marginals

Rethinking Neural-based Matrix Inversion: Why can't, and Where can

Double Debiased Machine Learning for Mediation Analysis with Continuous Treatments

Density Ratio-based Proxy Causal Learning Without Density Ratios

Learning Visual-Semantic Subspace Representations

On the Identifiability of Causal Abstractions

A primer on linear classification with missing data

When Can We Solve the Weighted Low Rank Approximation Problem in Truly Subquadratic Time?

Learning in Herding Mean Field Games: Single-Loop Algorithm with Finite-Time Convergence Analysis

Distributional Off-policy Evaluation with Bellman Residual Minimization

Generalized Criterion for Identifiability of Additive Noise Models Using Majorization

Safety in the Face of Adversity: Achieving Zero Constraint Violation in Online Learning with Slowly Changing Constraints

Consistent Amortized Clustering via Generative Flow Networks

Q-learning for Quantile MDPs: A Decomposition, Performance, and Convergence Analysis

Variational Inference on the Boolean Hypercube with the Quantum Entropy

Analyzing Generative Models by Manifold Entropic Metrics

Robust Gradient Descent for Phase Retrieval

MODL: Multilearner Online Deep Learning

Kernel Single Proxy Control for Deterministic Confounding

Unveiling the Role of Randomization in Multiclass Adversarial Classification: Insights from Graph Theory

Multimodal Learning with Uncertainty Quantification based on Discounted Belief Fusion

Nonparametric estimation of Hawkes processes with RKHSs

Contractivity and linear convergence in bilinear saddle-point problems: An operator-theoretic approach

Covariance Selection over Networks

Refined Analysis of Constant Step Size Federated Averaging and Federated Richardson-Romberg Extrapolation

High-Dimensional Differential Parameter Inference in Exponential Family using Time Score Matching

Domain Adaptation and Entanglement: an Optimal Transport Perspective

BudgetIV: Optimal Partial Identification of Causal Effects with Mostly Invalid Instruments

Fully Dynamic Adversarially Robust Correlation Clustering in Polylogarithmic Update Time

Improving N-Glycosylation and Biopharmaceutical Production Predictions Using AutoML-Built Residual Hybrid Models

Classification of High-dimensional Time Series in Spectral Domain Using Explainable Features with Applications to Neuroimaging Data

Meta-learning Task-specific Regularization Weights for Few-shot Linear Regression

Max-Rank: Efficient Multiple Testing for Conformal Prediction

LITE: Efficiently Estimating Gaussian Probability of Maximality

On the Difficulty of Constructing a Robust and Publicly-Detectable Watermark

SemlaFlow -- Efficient 3D Molecular Generation with Latent Attention and Equivariant Flow Matching

Understanding the Effect of GCN Convolutions in Regression Tasks

Large Covariance Matrix Estimation With Nonnegative Correlations

Diffusion Models as Constrained Samplers for Optimization with Unknown Constraints

On Tractability of Learning Bayesian Networks with Ancestral Constraints

Structure based SAT dataset for analysing GNN generalisation

Scalable Out-of-Distribution Robustness in the Presence of Unobserved Confounders

Transfer Learning for High-dimensional Reduced Rank Time Series Models

Continuous Structure Constraint Integration for Robust Causal Discovery

On the Relationship Between Robustness and Expressivity of Graph Neural Networks

Data Reconstruction Attacks and Defenses: A Systematic Evaluation

Decision from Suboptimal Classifiers: Excess Risk Pre- and Post-Calibration

Stein Boltzmann Sampling: A Variational Approach for Global Optimization

Asynchronous Decentralized Optimization with Constraints: Achievable Speeds of Convergence for Directed Graphs

Axiomatic Explainer Globalness via Optimal Transport

Optimistic Safety for Online Convex Optimization with Unknown Linear Constraints

Bridging Multiple Worlds: Multi-marginal Optimal Transport for Causal Partial-identification Problem

Approximate information maximization for bandit games

Understanding the Learning Dynamics of LoRA: A Gradient Flow Perspective on Low-Rank Adaptation in Matrix Factorization

Variational Combinatorial Sequential Monte Carlo for Bayesian Phylogenetics in Hyperbolic Space

Enhanced Adaptive Gradient Algorithms for Nonconvex-PL Minimax Optimization

Partial Information Decomposition for Data Interpretability and Feature Selection

Training Neural Samplers with Reverse Diffusive KL Divergence

Approximate Global Convergence of Independent Learning in Multi-Agent Systems

A Unified Evaluation Framework for Epistemic Predictions

Ant Colony Sampling with GFlowNets for Combinatorial Optimization

Calibrated Computation-Aware Gaussian Processes

Parameter estimation in state space models using particle importance sampling

When the Universe is Too Big: Bounding Consideration Probabilities for Plackett-Luce Rankings

Optimal estimation of linear non-Gaussian structure equation models

Graph-based Complexity for Causal Effect by Empirical Plug-in

Active Bipartite Ranking with Smooth Posterior Distributions

A graphical global optimization framework for parameter estimation of statistical models with nonconvex regularization functions

Variational Schr\"odinger Momentum Diffusion

Decoupling epistemic and aleatoric uncertainties with possibility theory

All models are wrong, some are useful: Model Selection with Limited Labels

From Gradient Clipping to Normalization for Heavy Tailed SGD

Computing high-dimensional optimal transport by flow neural networks

MING: A Functional Approach to Learning Molecular Generative Models

Sampling from Bayesian Neural Network Posteriors with Symmetric Minibatch Splitting Langevin Dynamics

Nonparametric Factor Analysis and Beyond

Locally Optimal Descent for Dynamic Stepsize Scheduling

Evidential Uncertainty Probes for Graph Neural Networks

Distance Estimation for High-Dimensional Discrete Distributions

Personalized Convolutional Dictionary Learning of Physiological Time Series

TRADE: Transfer of Distributions between External Conditions with Normalizing Flows

Signed Graph Autoencoder for Explainable and Polarization-Aware Network Embeddings

Flexible and Efficient Probabilistic PDE Solvers through Gaussian Markov Random Fields

Regularity in Canonicalized Models: A Theoretical Perspective

Causal Discovery on Dependent Binary Data

Spectral Differential Network Analysis for High-Dimensional Time Series

The VampPrior Mixture Model

Multi-Agent Credit Assignment with Pretrained Language Models

A Convex Relaxation Approach to Generalization Analysis for Parallel Positively Homogeneous Networks

Factor Analysis with Correlated Topic Model for Multi-Modal Data

Ordered $\mathcal{V}$-information Growth: A Fresh Perspective on Shared Information

Learning Gaussian Multi-Index Models with Gradient Flow: Time Complexity and Directional Convergence

$\mathcal{I}$-trustworthy Models. A framework for trustworthiness evaluation of probabilistic classifiers

An Iterative Algorithm for Rescaled Hyperbolic Functions Regression

Calm Composite Losses: Being Improper Yet Proper Composite

Credal Two-Sample Tests of Epistemic Uncertainty

ADEPT: Hierarchical Bayes Approach to Personalized Federated Unsupervised Learning

Adversarially-Robust TD Learning with Markovian Data: Finite-Time Rates and Fundamental Limits

Memory-Efficient Optimization with Factorized Hamiltonian Descent

Cross Validation for Correlated Data in Classification Models

Empirical Error Estimates for Graph Sparsification

Bayesian Inference in Recurrent Explicit Duration Switching Linear Dynamical Systems

Loss Gradient Gaussian Width based Generalization and Optimization Guarantees

Zero-Shot Action Generalization with Limited Observations

Theoretically Grounded Pruning of Large Ground Sets for Constrained, Discrete Optimization

Deep Optimal Sensor Placement for Black Box Stochastic Simulations

Truncated Inverse-Lévy Measure Representation of the Beta Process

On the Convergence of Continual Federated Learning Using Incrementally Aggregated Gradients

Improving Pre-trained Self-Supervised Embeddings Through Effective Entropy Maximization

A Unifying Framework for Action-Conditional Self-Predictive Reinforcement Learning

Perfect Recovery for Random Geometric Graph Matching with Shallow Graph Neural Networks

On the Inherent Privacy of Zeroth-Order Projected Gradient Descent

SINE: Scalable MPE Inference for Probabilistic Graphical Models using Advanced Neural Embeddings

Robust Kernel Hypothesis Testing under Data Corruption

Vecchia Gaussian Process Ensembles on Internal Representations of Deep Neural Networks

Parabolic Continual Learning

Tamed Langevin sampling under weaker conditions

A Differential Inclusion Approach for Learning Heterogeneous Sparsity in Neuroimaging Analysis

Learning Identifiable Structures Helps Avoid Bias in DNN-based Supervised Causal Learning

Beyond Discretization: Learning the Optimal Solution Path

Analysis of Two-Stage Rollout Designs with Clustering for Causal Inference under Network Interference

Energy-consistent Neural Operators for Hamiltonian and Dissipative Partial Differential Equations

Poisoning Bayesian Inference via Data Deletion and Replication

Survival Models: Proper Scoring Rule and Stochastic Optimization with Competing Risks

What Ails Generative Structure-based Drug Design: Expressivity is Too Little or Too Much?

Causal Temporal Regime Structure Learning

UNHaP: Unmixing Noise from Hawkes Processes

Disentangling impact of capacity, objective, batchsize, estimators, and step-size on flow VI

Model Evaluation in the Dark: Robust Classifier Metrics with Missing Labels

Fairness Risks for Group-Conditionally Missing Demographics

Copula Based Trainable Calibration Error Estimator of Multi-Label Classification with Label Interdependencies

Sampling from the Random Linear Model via Stochastic Localization Up to the AMP Threshold

Near-Polynomially Competitive Active Logistic Regression

The Strong Product Model for Network Inference without Independence Assumptions

Hyperboloid GPLVM for Discovering Continuous Hierarchies via Nonparametric Estimation

ScoreFusion: Fusing Score-based Generative Models via Kullback–Leibler Barycenters

Statistical Test for Auto Feature Engineering by Selective Inference

Disentangling Interactions and Dependencies in Feature Attributions

New User Event Prediction Through the Lens of Causal Inference

Adapting to Online Distribution Shifts in Deep Learning: A Black-Box Approach

Infinite Width Limits of Self Supervised Neural Networks

Gated Recurrent Neural Networks with Weighted Time-Delay Feedback

Scalable Implicit Graphon Learning

On Subjective Uncertainty Quantification and Calibration in Natural Language Generation

Is Gibbs sampling faster than Hamiltonian Monte Carlo on GLMs?

Revisiting LocalSGD and SCAFFOLD: Improved Rates and Missing Analysis

posteriordb: Testing, Benchmarking and Developing Bayesian Inference Algorithms

AxlePro: Momentum-Accelerated Batched Training of Kernel Machines

Tensor Network Based Feature Learning Model

Credibility-Aware Multimodal Fusion Using Probabilistic Circuits

A Family of Distributions of Random Subsets for Controlling Positive and Negative Dependence

Mean-Field Microcanonical Gradient Descent

Scalable Inference for Bayesian Multinomial Logistic-Normal Dynamic Linear Models

On the Geometry and Optimization of Polynomial Convolutional Networks

Riemann$^2$: Learning Riemannian Submanifolds from Riemannian Data

Generalization Lower Bounds for GD and SGD in Smooth Stochastic Convex Optimization

Posterior Mean Matching: Generative Modeling through Online Bayesian Inference

Implicit Diffusion: Efficient optimization through stochastic sampling

RetroDiff: Retrosynthesis as Multi-stage Distribution Interpolation

InnerThoughts: Disentangling Representations and Predictions in Large Language Models

Revisiting Online Learning Approach to Inverse Linear Optimization: A Fenchel–Young Loss Perspective and Gap-Dependent Regret Analysis

FedBaF: Federated Learning Aggregation Biased by a Foundation Model

Mixed-Feature Logistic Regression Robust to Distribution Shifts

Offline RL via Feature-Occupancy Gradient Ascent

Clustering Context in Off-Policy Evaluation

Synthetic Potential Outcomes and Causal Mixture Identifiability

Adaptive Extragradient Methods for Root-finding Problems under Relaxed Assumptions

Heterogeneous Graph Structure Learning through the Lens of Data-generating Processes

Integer Programming Based Methods and Heuristics for Causal Graph Learning

Unconditionally Calibrated Priors for Beta Mixture Density Networks

RTD-Lite: Scalable Topological Analysis for Comparing Weighted Graphs in Learning Tasks

Strong Screening Rules for Group-based SLOPE Models

Conditioning diffusion models by explicit forward-backward bridging

Separation-Based Distance Measures for Causal Graphs

HAVER: Instance-Dependent Error Bounds for Maximum Mean Estimation and Applications to Q-Learning and Monte Carlo Tree Search

Tight Analysis of Difference-of-Convex Algorithm (DCA) Improves Convergence Rates for Proximal Gradient Descent

FreqMoE: Enhancing Time Series Forecasting through Frequency Decomposition Mixture of Experts

Learning Geometrically-Informed Lyapunov Functions with Deep Diffeomorphic RBF Networks

Every Call is Precious: Global Optimization of Black-Box Functions with Unknown Lipschitz Constants

Stochastic Compositional Minimax Optimization with Provable Convergence Guarantees

Generalization Bounds for Dependent Data using Online-to-Batch Conversion.

Meta-learning from Heterogeneous Tensors for Few-shot Tensor Completion

Optimal Time Complexity Algorithms for Computing General Random Walk Graph Kernels on Sparse Graphs

Learning-Augmented Algorithms for Online Concave Packing and Convex Covering Problems

S-CFE: Simple Counterfactual Explanations

Rate of Model Collapse in Recursive Training

Robust Classification by Coupling Data Mollification with Label Smoothing

Learning to Forget: Bayesian Time Series Forecasting using Recurrent Sparse Spectrum Signature Gaussian Processes

Improving Stochastic Cubic Newton with Momentum

Personalizing Low-Rank Bayesian Neural Networks Via Federated Learning

(ends 4:00 AM)

6:30 p.m.

Registration Desk

(ends 2:00 AM)

7 p.m.

Invited Talk:

Keynote: Aapo Hyvärinen

Aapo Hyvarinen

(ends 8:00 PM)

8 p.m.

Break:

Coffee Break

(ends 8:30 PM)

8:30 p.m.

Oral Session 5: Probabilistic Inference and Optimzation [8:30-9:30]

Orals -

Entropic Matching for Expectation Propagation of Markov Jump Processes

Information Transfer Across Clinical Tasks via Adaptive Parameter Optimisation

posteriordb: Testing, Benchmarking and Developing Bayesian Inference Algorithms

Restructuring Tractable Probabilistic Circuits

Variation Due to Regularization Tractably Recovers Bayesian Deep Learning Uncertainty

(ends 9:30 PM)

9:30 p.m.

Oral Session 6: RL and Dynamical Systems [9:30-10:30]

Orals -

Corruption Robust Offline Reinforcement Learning with Human Feedback

Hybrid Transfer Reinforcement Learning: Provable Sample Efficiency from Shifted-Dynamics Data

Multi-marginal Schrödinger Bridges with Iterative Reference Refinement

Near-Optimal Algorithm for Non-Stationary Kernelized Bandits

Pure Exploration with Feedback Graphs

(ends 10:30 PM)

10:30 p.m.

Break:

Lunch + DEI Events

(ends 12:00 AM)

MON 5 MAY

midnight

Oral Session 7: Robust Learning [12:00-1:00]

Orals -

A Robust Kernel Statistical Test of Invariance: Detecting Subtle Asymmetries

Certifiably Quantisation-Robust training and inference of Neural Networks

Learning from biased positive-unlabeled data via threshold calibration

Robust Kernel Hypothesis Testing under Data Corruption

Statistical Learning of Distributionally Robust Stochastic Control in Continuous State Spaces

(ends 1:00 AM)

1 a.m.

Poster Session 3 [1:00-4:00]

Posters -

Pick-to-Learn and Self-Certified Gaussian Process Approximations

Additive Model Boosting: New Insights and Path(ologie)s

Do Regularization Methods for Shortcut Mitigation Work As Intended?

Diffusion Models under Group Transformations

StableMDS: A Novel Gradient Descent-Based Method for Stabilizing and Accelerating Weighted Multidimensional Scaling

Transfer Neyman-Pearson Algorithm for Outlier Detection

$\beta$-th order Acyclicity Derivatives for DAG Learning

Powerful batch conformal prediction for classification

Near-Optimal Sample Complexity for Iterated CVaR Reinforcement Learning with a Generative Model

Sparse Activations as Conformal Predictors

Optimising Clinical Federated Learning through Mode Connectivity-based Model Aggregation

Safe exploration in reproducing kernel Hilbert spaces

Certifiably Quantisation-Robust training and inference of Neural Networks

Geometry-Aware Generative Autoencoders for Warped Riemannian Metric Learning and Generative Modeling on Data Manifolds

Your copula is a classifier in disguise: classification-based copula density estimation

Graph Machine Learning based Doubly Robust Estimator for Network Causal Effects

Feasible Learning

Order-Optimal Regret with Novel Policy Gradient Approaches in Infinite-Horizon Average Reward MDPs

Subspace Recovery in Winsorized PCA: Insights into Accuracy and Robustness

Efficient Optimization Algorithms for Linear Adversarial Training

Protein Fitness Landscape: Spectral Graph Theory Perspective

Policy Teaching via Data Poisoning in Learning from Human Preferences

Superiority of Multi-Head Attention: A Theoretical Study in Shallow Transformers in In-Context Linear Regression

Restructuring Tractable Probabilistic Circuits

Beyond Size-Based Metrics: Measuring Task-Specific Complexity in Symbolic Regression

Differentially Private Kernelized Contextual Bandits

Task Shift: From Classification to Regression in Overparameterized Linear Models

Hyperbolic Prototypical Entailment Cones for Image Classification

Natural Language Counterfactual Explanations for Graphs Using Large Language Models

Paths and Ambient Spaces in Neural Loss Landscapes

Wasserstein Distributionally Robust Bayesian Optimization with Continuous Context

Global Group Fairness in Federated Learning via Function Tracking

A Computation-Efficient Method of Measuring Dataset Quality based on the Coverage of the Dataset

Theory of Agreement-on-the-Line in Linear Models and Gaussian Data

A Robust Kernel Statistical Test of Invariance: Detecting Subtle Asymmetries

Adaptive Convergence Rates for Log-Concave Maximum Likelihood

Bayesian Decision Theory on Decision Trees: Uncertainty Evaluation and Interpretability

On Tradeoffs in Learning-Augmented Algorithms

Accuracy on the wrong line: On the pitfalls of noisy data for out-of-distribution generalisation

On the Sample Complexity of Next-Token Prediction

On adaptivity and minimax optimality of two-sided nearest neighbors

Quantile Additive Trend Filtering

Synthesis and Analysis of Data as Probability Measures With Entropy-Regularized Optimal Transport

Advancing Fairness in Precision Medicine: A Universal Framework for Optimal Treatment Estimation in Censored Data

Primal-Dual Spectral Representation for Off-policy Evaluation

Pure Exploration with Feedback Graphs

On the Power of Multitask Representation Learning with Gradient Descent

A Safe Exploration Approach to Constrained Markov Decision Processes

Theoretical Convergence Guarantees for Variational Autoencoders

Sampling From Multiscale Densities With Delayed Rejection Generalized Hamiltonian Monte Carlo

Lower Bounds for Time-Varying Kernelized Bandits

Efficient and Asymptotically Unbiased Constrained Decoding for Large Language Models

Decision-Point Guided Safe Policy Improvement

Reward Maximization for Pure Exploration: Minimax Optimal Good Arm Identification for Nonparametric Multi-Armed Bandits

High-probability Convergence Bounds for Online Nonlinear Stochastic Gradient Descent under Heavy-tailed Noise

Multi-level Advantage Credit Assignment for Cooperative Multi-Agent Reinforcement Learning

To Give or Not to Give? The Impacts of Strategically Withheld Recourse

Representer Theorems for Metric and Preference Learning: Geometric Insights and Algorithms

Computation-Aware Kalman Filtering and Smoothing

Understanding GNNs and Homophily in Dynamic Node Classification

Nyström Kernel Stein Discrepancy

Visualizing token importance for black-box language models

Sparse Causal Effect Estimation using Two-Sample Summary Statistics in the Presence of Unmeasured Confounding

Federated UCBVI: Communication-Efficient Federated Regret Minimization with Heterogeneous Agents

Understanding Inverse Reinforcement Learning under Overparameterization: Non-Asymptotic Analysis and Global Optimality

Estimating the Spectral Moments of the Kernel Integral Operator from Finite Sample Matrices

Active Feature Acquisition for Personalised Treatment Assignment

Federated Causal Inference: Multi-Study ATE Estimation beyond Meta-Analysis

Trustworthy assessment of heterogeneous treatment effect estimator via analysis of relative error

Koopman-Equivariant Gaussian Processes

Learning signals defined on graphs with optimal transport and Gaussian process regression

Statistical Guarantees for Unpaired Image-to-Image Cross-Domain Analysis using GANs

Infinite-Horizon Reinforcement Learning with Multinomial Logit Function Approximation

Conditional simulation via entropic optimal transport: Toward non-parametric estimation of conditional Brenier maps

Geometric Collaborative Filtering with Convergence

Noise-Aware Differentially Private Variational Inference

The Uniformly Rotated Mondrian Kernel

Logarithmic Neyman Regret for Adaptive Estimation of the Average Treatment Effect

The Local Learning Coefficient: A Singularity-Aware Complexity Measure

Collaborative non-parametric two-sample testing

Learning High-dimensional Gaussians from Censored Data

Clustered Invariant Risk Minimization

Sequential Kernelized Stein Discrepancy

Differentially Private Graph Data Release: Inefficiencies & Unfairness

DDEQs: Distributional Deep Equilibrium Models through Wasserstein Gradient Flows

HR-Bandit: Human-AI Collaborated Linear Recourse Bandit

Learning Laplacian Positional Encodings for Heterophilous Graphs

Inverse Optimization with Prediction Market: A Characterization of Scoring Rules for Elciting System States

A Subquadratic Time Approximation Algorithm for Individually Fair k-Center

Is Merging Worth It? Securely Evaluating the Information Gain for Causal Dataset Acquisition

Weighted Sum of Gaussian Process Latent Variable Models

Best-Arm Identification in Unimodal Bandits

TVineSynth: A Truncated C-Vine Copula Generator of Synthetic Tabular Data to Balance Privacy and Utility

Conditional diffusions for amortized neural posterior estimation

Efficient Trajectory Inference in Wasserstein Space Using Consecutive Averaging

Density Ratio Estimation via Sampling along Generalized Geodesics on Statistical Manifolds

On the Consistent Recovery of Joint Distributions from Conditionals

Stochastic Gradient Descent for Bézier Simplex Representation of Pareto Set in Multi-Objective Optimization

Score matching for bridges without learning time-reversals

$f$-PO: Generalizing Preference Optimization with $f$-divergence Minimization

Reliable and Scalable Variable Importance Estimation via Warm-start and Early Stopping

Elastic Representation: Mitigating Spurious Correlations for Group Robustness

Quantifying Knowledge Distillation using Partial Information Decomposition

Understanding Expert Structures on Minimax Parameter Estimation in Contaminated Mixture of Experts

Reinforcement Learning for Infinite-Horizon Average-Reward Linear MDPs via Approximation by Discounted-Reward MDPs

Prediction-Centric Uncertainty Quantification via MMD

Robust Multi-fidelity Bayesian Optimization with Deep Kernel and Partition

Learning the Pareto Front Using Bootstrapped Observation Samples

Entropic Matching for Expectation Propagation of Markov Jump Processes

DeCaf: A Causal Decoupling Framework for OOD Generalization on Node Classification

Variance-Aware Linear UCB with Deep Representation for Neural Contextual Bandits

Tensor Network-Constrained Kernel Machines as Gaussian Processes

Achieving $\widetilde{\mathcal{O}}(\sqrt{T})$ Regret in Average-Reward POMDPs with Known Observation Models

Stochastic Rounding for LLM Training: Theory and Practice

Causal Discovery-Driven Change Point Detection in Time Series

Bayesian Circular Regression with von Mises Quasi-Processes

Adversarial Vulnerabilities in Large Language Models for Time Series Forecasting

Anytime-Valid A/B Testing of Counting Processes

Task-Driven Discrete Representation Learning

Corruption Robust Offline Reinforcement Learning with Human Feedback

Algorithmic Accountability in Small Data: Sample-Size-Induced Bias Within Classification Metrics

Testing Conditional Independence with Deep Neural Network Based Binary Expansion Testing (DeepBET)

$q\texttt{POTS}$: Efficient Batch Multiobjective Bayesian Optimization via Pareto Optimal Thompson Sampling

Consistent Validation for Predictive Methods in Spatial Settings

Models That Are Interpretable But Not Transparent

Get rid of your constraints and reparametrize: A study in NNLS and implicit bias

Composition and Control with Distilled Energy Diffusion Models and Sequential Monte Carlo

Permutation Invariant Functions: Statistical Testing, Density Estimation, and Metric Entropy

Q-function Decomposition with Intervention Semantics for Factored Action Spaces

Differentially Private Range Queries with Correlated Input Perturbation

Cubic regularized subspace Newton for non-convex optimization

The cost of local and global fairness in Federated Learning

Differentiable Calibration of Inexact Stochastic Simulation Models via Kernel Score Minimization

Black-Box Uniform Stability for Non-Euclidean Empirical Risk Minimization

M-HOF-Opt: Multi-Objective Hierarchical Output Feedback Optimization via Multiplier Induced Loss Landscape Scheduling

Data-Driven Upper Confidence Bounds with Near-Optimal Regret for Heavy-Tailed Bandits

Bilevel Reinforcement Learning via the Development of Hyper-gradient without Lower-Level Convexity

Adversarial Training in High-Dimensional Regression: Generated Data and Neural Networks

On the Power of Adaptive Weighted Aggregation in Heterogeneous Federated Learning and Beyond

Parallel Backpropagation for Inverse of a Convolution with Application to Normalizing Flows

Performative Reinforcement Learning with Linear Markov Decision Process

A Novel Convex Gaussian Min Max Theorem for Repeated Features

What and How does In-Context Learning Learn? Bayesian Model Averaging, Parameterization, and Generalization

Constrained Multi-objective Bayesian Optimization through Optimistic Constraints Estimation

Weighted Euclidean Distance Matrices over Mixed Continuous and Categorical Inputs for Gaussian Process Models

From Deep Additive Kernel Learning to Last-Layer Bayesian Neural Networks via Induced Prior Approximation

ClusterSC: Advancing Synthetic Control with Donor Selection

Two-Timescale Linear Stochastic Approximation: Constant Stepsizes Go a Long Way

A Multi-Armed Bandit Approach to Online Selection and Evaluation of Generative Models

Robust Offline Policy Learning with Observational Data from Multiple Sources

Near-Optimal Sample Complexity in Reward-Free Kernel-based Reinforcement Learning

Enhancing Feature-Specific Data Protection via Bayesian Coordinate Differential Privacy

Learning from biased positive-unlabeled data via threshold calibration

Semiparametric conformal prediction

A Theoretical Framework for Preventing Class Collapse in Supervised Contrastive Learning

M$^2$AD: Multi-Sensor Multi-System Anomaly Detection through Global Scoring and Calibrated Thresholding

Selecting the Number of Communities for Weighted Degree-Corrected Stochastic Block Models

Robust Fair Clustering with Group Membership Uncertainty Sets

Density-Dependent Group Testing

Post-processing for Fair Regression via Explainable SVD

No-Regret Bayesian Optimization with Stochastic Observation Failures

Multi-agent Multi-armed Bandit Regret Complexity and Optimality

Invertible Fourier Neural Operators for Tackling Both Forward and Inverse Problems

On Distributional Discrepancy for Experimental Design with General Assignment Probabilities

Learning Pareto manifolds in high dimensions: How can regularization help?

High Dimensional Bayesian Optimization using Lasso Variable Selection

Global Optimization of Gaussian Process Acquisition Functions Using a Piecewise-Linear Kernel Approximation

ROTI-GCV: Generalized Cross-Validation for right-ROTationally Invariant Data

LC-Tsallis-INF: Generalized Best-of-Both-Worlds Linear Contextual Bandits

Conformal Prediction Under Generalized Covariate Shift with Posterior Drift

Accelerated Methods for Riemannian Min-Max Optimization Ensuring Bounded Geometric Penalties

Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models

Strategic Conformal Prediction

Change Point Detection in Hadamard Spaces by Alternating Minimization

Information Transfer Across Clinical Tasks via Adaptive Parameter Optimisation

The Hardness of Validating Observational Studies with Experimental Data

Linearized Wasserstein Barycenters: Synthesis, Analysis, Representational Capacity, and Applications

Harnessing Causality in Reinforcement Learning with Bagged Decision Times

Cost-Aware Optimal Pairwise Pure Exploration

Online Student-$t$ Processes with an Overall-local Scale Structure for Modelling Non-stationary Data

Stochastic Weight Sharing for Bayesian Neural Networks

Leveraging Frozen Batch Normalization for Co-Training in Source-Free Domain Adaptation

Neural Point Processes for Pixel-wise Regression

Dynamic DBSCAN with Euler Tour Sequences

From Learning to Optimize to Learning Optimization Algorithms

The Pivoting Framework: Frank-Wolfe Algorithms with Active Set Size Control

Explaining ViTs Using Information Flow

Dissecting the Impact of Model Misspecification in Data-Driven Optimization

Performative Prediction on Games and Mechanism Design

A Shapley-value Guided Rationale Editor for Rationale Learning

Noisy Low-Rank Matrix Completion via Transformed $L_1$ Regularization and its Theoretical Properties

Causal Representation Learning from General Environments under Nonparametric Mixing

Approximate Equivariance in Reinforcement Learning

Towards a mathematical theory for consistency training in diffusion models

Nonparametric Distributional Regression via Quantile Regression

(ends 4:00 AM)