All papers

Develops robust estimators and optimization algorithms to learn linear dynamical systems from trajectories corrupted by adversarial outliers.

Proposes a market-level mixture-of-experts coordination framework using cost-aware gating and revenue allocation across heterogeneous model services.

Presents a probabilistic framework to disaggregate coarse socioeconomic observations into fine spatial scales using satellite embeddings.

Introduces a continuous-time causal prior-data fitted network simulator to estimate market impact without confounding biases.

Provides theoretical guarantees for efficiently learning Cox survival analysis models using first-order optimization under outcome-dependent truncation and censoring.

Develops a low-cost piezoresistive glove system that aligns human touch demonstrations with robot sensors via contact events.

Presents a score-based method to learn cluster causal directed acyclic graphs from interventional data under linear Gaussian assumptions.

Introduces an imprecise Dirichlet model method to calculate interval-valued Shapley values and analyze attribution robustness in tree-based models.

Proposes a reliability-aware conditioning method that mitigates harmful temporal misalignments when using generated video futures for robot manipulation.

Introduces an adaptation framework for frozen vision-language-action policies using residual reinforcement learning and online capability inference during joint faults.

Presents scalable methods using distance sketches and random feature embeddings to efficiently approximate quadratic entropy across large distributions.

Learns linear shape parameterisations from past designs using PCA to improve sample efficiency in Bayesian shape optimisation.

Analyzes test-time compute strategies, including parameter adaptation and prediction aggregation, to improve predictions in tabular foundation models.

The paper proposes ghost tasking, adding auxiliary tasks to enable parametrized Gaussian processes for non-parametrizable linear differential equations.

The paper introduces CausalDreamer, disentangling world model latent representations along controllability and reward relevance dimensions for control.

The paper develops an active-set likelihood framework for archetypal analysis on discrete binary, count, and categorical data.

The paper analyzes Gaussian diffusion sampling to show how time-dependent trajectories eliminate condition number dependence seen in Langevin dynamics.

The paper combines holomorphic neural networks with singular basis functions to solve 2D and 3D Laplace boundary-value problems.

The paper proposes using simulator output gradients within Bayesian active learning to infer input parameters of expensive differential equation simulators.

The paper introduces Differentiable Systematic Resampling, a smooth relaxation enabling gradient-based optimization in variational sequential Monte Carlo.

The paper introduces SCORE, an efficient spectral framework for approximating and estimating high-dimensional structured Gaussian covariance matrices.

The paper proposes a credal machine learning framework that models epistemic uncertainty with probability distribution sets for risk-averse decision-making.

The paper develops a neural density functional preserving physical symmetries to model ion correlations in inhomogeneous polymer electrolytes.

The paper evaluates 1,368 quantum neural network configurations on financial time series forecasting using spot exchange rate data.

The paper proposes a Bayesian optimization approach that uses linear response approximations to maintain state-preservation constraints in PDE systems.

The paper introduces MA-OPS to achieve optimal regret in adversarial Markov decision processes subject to stochastic hard constraints.

The paper investigates synthetic graph generation and introduces TS-GGAD to train generalist graph anomaly detection models without real-world training graphs.

The paper proposes DI-SCUSUM, a training-free quickest change detection method using diffusion-smoothed empirical score differences.

The paper proposes ISBO, combining log Gaussian Cox processes with the INLA-SPDE approach for scalable spatio-temporal Bayesian optimization.

The paper audits test-time adaptation methods for time-series forecasting under realistic, delay-aware conditions to evaluate actual streaming performance.

The paper introduces RIFT, a deterministic, parameter-free anomaly detection algorithm based on minimum spanning trees inspired by Isolation Forests.

The paper presents AdaptLSTM, an adaptive online learning method that selectively updates cloud workload forecasts upon detecting distribution drift.

The paper examines how multi-agent Q-learning movement policies generate spatial clustering and traveling bands in spatial public goods games.

The paper develops an online-to-nonconvex conversion framework for stochastic nonsmooth nonconvex composite optimization achieving optimal oracle complexity bounds.

The paper presents asdex, an automatic sparse differentiation library in JAX that uses graph coloring to compute sparse Jacobians efficiently.

The paper introduces RefineMix to train discrete diffusion models under data scarcity using out-of-distribution data at chosen diffusion timesteps.

The paper presents SplitJEPA, learning invariant and variant latent world state representations directly without using reconstruction objectives.

This paper derives sharp sample complexity thresholds for Gaussian regression over bounded-degree polynomial subspaces on the Boolean cube.

This paper establishes tighter regret bounds for adversarial Markov decision processes by incorporating regularized Q-functions into policy optimization.

This paper uses bilevel optimization and kernel autoencoders to learn finite-dimensional Koopman operator embeddings from nonlinear dynamical data.

This paper presents a transformer architecture that predicts conditional marginal distributions for missing data in a single forward pass.

This paper analyzes source-side training dynamics to prospectively detect out-of-distribution performance degradation before evaluation.

This paper introduces a regional weather forecasting framework that aligns pretrained global model guidance with high-resolution regional representations.

This paper develops a unified Bellman operator to optimize task performance and enforce constraints within safety-critical reinforcement learning.

This paper provides statistical sample complexity guarantees for transformer-based masked diffusion models learning distributions with latent tree structures.

This paper investigates the relationship between circuit depth, training dataset size, and generalization bounds in adaptively grown quantum classifiers.

This paper presents a feasibility-aware safety filter that amortizes minimal-intervention constraints using a feedforward neural network in reinforcement learning.

This paper trains a generalist humanoid policy by predicting body and hand motion latents mapped from human demonstration data.

This paper estimates density ratios by parameterizing distribution shift as a displacement field using the Stein operator.

This paper redesigns 4-bit AdamW optimizer-state quantization by performing stochastic rounding directly within preconditioner space.

This paper analyzes data sampling distributions across parallel simulations to improve exploration efficiency in reinforcement learning for robot control.

This paper introduces a training-free safety filter that evaluates scene context to constrain generated humanoid motion trajectories.

This paper proposes a sample-adaptive zeroth-order optimization method that estimates sample importance online to accelerate large language model fine-tuning.

This paper designs and simulates an all-transistor physical circuit using sub-threshold MOSFETs to run transformer attention and in-context learning.

This paper proposes a reward aggregation method to stabilize and scale reinforcement learning fine-tuning for diffusion models.

This paper introduces a benchmark and a diffusion aggregation algorithm to evaluate and balance accuracy and realism in event sequence forecasting.

This paper presents a method for upcycling pretrained language models into sparse architectures using token-indexed embeddings in residual pathways.

This paper proposes an active learning framework that selectively requests expert feedback at high-value decision steps during LLM agent workflows.

This paper applies a convolutional Gaussian neural process and genetic optimization to select optimal aerosol monitoring sensor locations across the UK.

This paper introduces a one-step implicit sampler using short-time MCMC dynamics to enable efficient amortized inference for Bayesian posteriors.