Andrej Risteski

Steering diffusion models with quadratic rewards: a fine-grained analysis COLT 2026 On the Benefits of Memory for Modeling Time-Dependent PDEs ICLR 2025 On the Query Complexity of Verifier-Assisted Language Generation ICML 2025 Progressive distillation induces an implicit curriculum ICLR 2025 Towards characterizing the value of edge embeddings in Graph Neural Networks ICML 2025 Fit Like You Sample: Sample-Efficient Generalized Score Matching from Fast Mixing Diffusions COLT 2024 Outliers with Opposing Signals Have an Outsized Effect on Neural Network Optimization ICLR 2024 Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines ICML 2024 Understanding Augmentation-based Self-Supervised Representation Learning via RKHS Approximation and Regression ICLR 2024 Deep Equilibrium Based Neural Operators for Steady-State PDEs NeurIPS 2023 How Do Transformers Learn Topic Structure: Towards a Mechanistic Understanding ICML 2023 Neural Network Approximations of PDEs Beyond Linearity: A Representational Perspective ICML 2023 Pitfalls of Gaussians as a noise distribution in NCE ICLR 2023 Provable benefits of annealing for estimating normalizing constants: Importance Sampling, Noise-Contrastive Estimation, and beyond NeurIPS 2023 Provable benefits of score matching NeurIPS 2023 Statistical Efficiency of Score Matching: The View from Isoperimetry ICLR 2023 Transformers are uninterpretable with myopic methods: a case study with bounded Dyck grammars NeurIPS 2023 Analyzing and Improving the Optimization Landscape of Noise-Contrastive Estimation ICLR 2022 An Online Learning Approach to Interpolation and Extrapolation in Domain Generalization AISTATS 2022 Continual learning: a feature extraction formalization, an efficient algorithm, and fundamental obstructions NeurIPS 2022 Contrasting the landscape of contrastive and non-contrastive learning AISTATS 2022 Iterative Feature Matching: Toward Provable Domain Generalization with Logarithmic Environments NeurIPS 2022 Masked Prediction: A Parameter Identifiability View NeurIPS 2022 Sampling Approximately Low-Rank Ising Models: MCMC meets Variational Methods COLT 2022 The Effects of Invertibility on the Representational Complexity of Encoders in Variational Autoencoders ICLR 2022 Variational autoencoders in the presence of low-dimensional data: landscape and implicit bias ICLR 2022 Contrastive learning of strong-mixing continuous-time stochastic processes AISTATS 2021 Efficient sampling from the Bingham distribution ALT 2021 Parametric Complexity Bounds for Approximating PDEs with Neural Networks NeurIPS 2021 Representational aspects of depth and conditioning in normalizing flows ICML 2021 The Limitations of Limited Context for Constituency Parsing ACL 2021 The Risks of Invariant Risk Minimization ICLR 2021 Universal Approximation Using Well-Conditioned Normalizing Flows NeurIPS 2021 Empirical Study of the Benefits of Overparameterization in Learning Latent Variable Models ICML 2020 On Learning Language-Invariant Representations for Universal Machine Translation ICML 2020 Approximability of Discriminators Implies Diversity in GANs ICLR 2019 Sum-of-squares meets square loss: Fast rates for agnostic tensor completion COLT 2019 The Comparative Power of ReLU Networks and Polynomial Kernels in the Presence of Sparse Latent Structure ICLR 2019 Beyond Log-concavity: Provable Guarantees for Sampling Multi-modal Distributions using Simulated Tempering Langevin Monte Carlo NeurIPS 2018 Do GANs learn the distribution? Some Theory and Empirics ICLR 2018 On the Ability of Neural Nets to Express Distributions COLT 2017 Algorithms and matching lower bounds for approximately-convex optimization NeurIPS 2016 Approximate maximum entropy principles via Goemans-Williamson with applications to provable variational methods NeurIPS 2016 How to calculate partition functions using convex programming hierarchies: provable bounds for variational methods COLT 2016 Recovery Guarantee of Non-negative Matrix Factorization via Alternating Updates NeurIPS 2016 Recovery guarantee of weighted low-rank approximation via alternating minimization ICML 2016 Label optimal regret bounds for online local learning COLT 2015 On some provably correct cases of variational inference for topic models NeurIPS 2015