Andrej Risteski

117 papers A* 37A 3B 1Journal 65Unranked 10

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	CodePDE: An Inference Framework for LLM-driven PDE Solver Generation. Trans. Mach. Learn. Res. Shanda Li, Tanya Marwah, Junhong Shen, Weiwei Sun, Andrej Risteski, Yiming Yang, Ameet Talwalkar
2026	J	`jnl`	Steering diffusion models with quadratic rewards: a fine-grained analysis. CoRR Ankur Moitra, Andrej Risteski, Dhruv Rohatgi
2025	J	`jnl`	CodePDE: An Inference Framework for LLM-driven PDE Solver Generation. CoRR Shanda Li, Tanya Marwah, Junhong Shen, Weiwei Sun, Andrej Risteski, Yiming Yang, Ameet Talwalkar
2025	A*	`conf`	On the Benefits of Memory for Modeling Time-Dependent PDEs. ICLR Ricardo Buitrago Ruiz, Tanya Marwah, Albert Gu, Andrej Risteski
2025	A*	`conf`	On the Query Complexity of Verifier-Assisted Language Generation. ICML Edoardo Botta, Yuchen Li, Aashay Mehta, Jordan T. Ash, Cyril Zhang, Andrej Risteski
2025	J	`jnl`	On the Query Complexity of Verifier-Assisted Language Generation. CoRR Edoardo Botta, Yuchen Li, Aashay Mehta, Jordan T. Ash, Cyril Zhang, Andrej Risteski
2025	A*	`conf`	Progressive distillation induces an implicit curriculum. ICLR Abhishek Panigrahi, Bingbin Liu, Sadhika Malladi, Andrej Risteski, Surbhi Goel
2025	J	`jnl`	Taming Imperfect Process Verifiers: A Sampling Perspective on Backtracking. CoRR Dhruv Rohatgi, Abhishek Shetty, Donya Saless, Yuchen Li, Ankur Moitra, Andrej Risteski, Dylan J. Foster
2025	J	`jnl`	Toward Understanding the Transferability of Adversarial Suffixes in Large Language Models. CoRR Sarah Ball, Niki Hasrati, Alexander Robey, Avi Schwarzschild, Frauke Kreuter, Zico Kolter, Andrej Risteski
2025	A*	`conf`	Towards characterizing the value of edge embeddings in Graph Neural Networks. ICML Dhruv Rohatgi, Tanya Marwah, Zachary Chase Lipton, Jianfeng Lu, Ankur Moitra, Andrej Risteski
2024	A*	`conf`	Fit Like You Sample: Sample-Efficient Generalized Score Matching from Fast Mixing Diffusions. COLT Yilong Qin, Andrej Risteski
2024	J	`jnl`	On the Benefits of Memory for Modeling Time-Dependent PDEs. CoRR Ricardo Buitrago Ruiz, Tanya Marwah, Albert Gu, Andrej Risteski
2024	A*	`conf`	Outliers with Opposing Signals Have an Outsized Effect on Neural Network Optimization. ICLR Elan Rosenfeld, Andrej Risteski
2024	J	`jnl`	Progressive distillation induces an implicit curriculum. CoRR Abhishek Panigrahi, Bingbin Liu, Sadhika Malladi, Andrej Risteski, Surbhi Goel
2024	A*	`conf`	Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines. ICML Yuchen Li, Alexandre Kirchmeyer, Aashay Mehta, Yilong Qin, Boris Dadachev, Kishore Papineni, Sanjiv Kumar, Andrej Risteski
2024	J	`jnl`	Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines. CoRR Yuchen Li, Alexandre Kirchmeyer, Aashay Mehta, Yilong Qin, Boris Dadachev, Kishore Papineni, Sanjiv Kumar, Andrej Risteski
2024	J	`jnl`	Towards characterizing the value of edge embeddings in Graph Neural Networks. CoRR Dhruv Rohatgi, Tanya Marwah, Zachary Chase Lipton, Jianfeng Lu, Ankur Moitra, Andrej Risteski
2024	A*	`conf`	Understanding Augmentation-based Self-Supervised Representation Learning via RKHS Approximation and Regression. ICLR Runtian Zhai, Bingbin Liu, Andrej Risteski, J. Zico Kolter, Pradeep Kumar Ravikumar
2023	A*	`conf`	Deep Equilibrium Based Neural Operators for Steady-State PDEs. NeurIPS Tanya Marwah, Ashwini Pokle, J. Zico Kolter, Zachary C. Lipton, Jianfeng Lu, Andrej Risteski
2023	J	`jnl`	Deep Equilibrium Based Neural Operators for Steady-State PDEs. CoRR Tanya Marwah, Ashwini Pokle, J. Zico Kolter, Zachary C. Lipton, Jianfeng Lu, Andrej Risteski
2023	J	`jnl`	Fit Like You Sample: Sample-Efficient Generalized Score Matching from Fast Mixing Markov Chains. CoRR Yilong Qin, Andrej Risteski
2023	A*	`conf`	How Do Transformers Learn Topic Structure: Towards a Mechanistic Understanding. ICML Yuchen Li, Yuanzhi Li, Andrej Risteski
2023	J	`jnl`	How Do Transformers Learn Topic Structure: Towards a Mechanistic Understanding. CoRR Yuchen Li, Yuanzhi Li, Andrej Risteski
2023	A*	`conf`	Neural Network Approximations of PDEs Beyond Linearity: A Representational Perspective. ICML Tanya Marwah, Zachary Chase Lipton, Jianfeng Lu, Andrej Risteski
2023	J	`jnl`	Outliers with Opposing Signals Have an Outsized Effect on Neural Network Optimization. CoRR Elan Rosenfeld, Andrej Risteski
2023	A*	`conf`	Pitfalls of Gaussians as a noise distribution in NCE. ICLR Holden Lee, Chirag Pabbaraju, Anish Prasad Sevekari, Andrej Risteski
2023	A*	`conf`	Provable benefits of annealing for estimating normalizing constants: Importance Sampling, Noise-Contrastive Estimation, and beyond. NeurIPS Omar Chehab, Aapo Hyvärinen, Andrej Risteski
2023	J	`jnl`	Provable benefits of annealing for estimating normalizing constants: Importance Sampling, Noise-Contrastive Estimation, and beyond. CoRR Omar Chehab, Aapo Hyvärinen, Andrej Risteski
2023	A*	`conf`	Provable benefits of score matching. NeurIPS Chirag Pabbaraju, Dhruv Rohatgi, Anish Prasad Sevekari, Holden Lee, Ankur Moitra, Andrej Risteski
2023	J	`jnl`	Provable benefits of score matching. CoRR Chirag Pabbaraju, Dhruv Rohatgi, Anish Prasad Sevekari, Holden Lee, Ankur Moitra, Andrej Risteski
2023	A*	`conf`	Statistical Efficiency of Score Matching: The View from Isoperimetry. ICLR Frederic Koehler, Alexander Heckett, Andrej Risteski
2023	A*	`conf`	Transformers are uninterpretable with myopic methods: a case study with bounded Dyck grammars. NeurIPS Kaiyue Wen, Yuchen Li, Bingbin Liu, Andrej Risteski
2023	J	`jnl`	Transformers are uninterpretable with myopic methods: a case study with bounded Dyck grammars. CoRR Kaiyue Wen, Yuchen Li, Bingbin Liu, Andrej Risteski
2023	J	`jnl`	Understanding Augmentation-based Self-Supervised Representation Learning via RKHS Approximation. CoRR Runtian Zhai, Bingbin Liu, Andrej Risteski, Zico Kolter, Pradeep Ravikumar
2022	A	`conf`	An Online Learning Approach to Interpolation and Extrapolation in Domain Generalization. AISTATS Elan Rosenfeld, Pradeep Ravikumar, Andrej Risteski
2022	A*	`conf`	Analyzing and Improving the Optimization Landscape of Noise-Contrastive Estimation. ICLR Bingbin Liu, Elan Rosenfeld, Pradeep Kumar Ravikumar, Andrej Risteski
2022	A*	`conf`	Continual learning: a feature extraction formalization, an efficient algorithm, and fundamental obstructions. NeurIPS Binghui Peng, Andrej Risteski
2022	J	`jnl`	Continual learning: a feature extraction formalization, an efficient algorithm, and fundamental obstructions. CoRR Binghui Peng, Andrej Risteski
2022	A	`conf`	Contrasting the landscape of contrastive and non-contrastive learning. AISTATS Ashwini Pokle, Jinjin Tian, Yuchen Li, Andrej Risteski
2022	J	`jnl`	Contrasting the landscape of contrastive and non-contrastive learning. CoRR Ashwini Pokle, Jinjin Tian, Yuchen Li, Andrej Risteski
2022	J	`jnl`	Domain-Adjusted Regression or: ERM May Already Learn Features Sufficient for Out-of-Distribution Generalization. CoRR Elan Rosenfeld, Pradeep Ravikumar, Andrej Risteski
2022	A*	`conf`	Iterative Feature Matching: Toward Provable Domain Generalization with Logarithmic Environments. NeurIPS Yining Chen, Elan Rosenfeld, Mark Sellke, Tengyu Ma, Andrej Risteski
2022	A*	`conf`	Masked Prediction: A Parameter Identifiability View. NeurIPS Bingbin Liu, Daniel J. Hsu, Pradeep Ravikumar, Andrej Risteski
2022	J	`jnl`	Masked prediction tasks: a parameter identifiability view. CoRR Bingbin Liu, Daniel Hsu, Pradeep Ravikumar, Andrej Risteski
2022	J	`jnl`	Neural Network Approximations of PDEs Beyond Linearity: Representational Perspective. CoRR Tanya Marwah, Zachary C. Lipton, Jianfeng Lu, Andrej Risteski
2022	J	`jnl`	Pitfalls of Gaussians as a noise distribution in NCE. CoRR Holden Lee, Chirag Pabbaraju, Anish Prasad Sevekari, Andrej Risteski
2022	A*	`conf`	Sampling Approximately Low-Rank Ising Models: MCMC meets Variational Methods. COLT Frederic Koehler, Holden Lee, Andrej Risteski
2022	J	`jnl`	Sampling Approximately Low-Rank Ising Models: MCMC meets Variational Methods. CoRR Frederic Koehler, Holden Lee, Andrej Risteski
2022	J	`jnl`	Statistical Efficiency of Score Matching: The View from Isoperimetry. CoRR Frederic Koehler, Alexander Heckett, Andrej Risteski
2022	A*	`conf`	The Effects of Invertibility on the Representational Complexity of Encoders in Variational Autoencoders. ICLR Divyansh Pareek, Andrej Risteski
2022	A*	`conf`	Variational autoencoders in the presence of low-dimensional data: landscape and implicit bias. ICLR Frederic Koehler, Viraj Mehta, Chenghui Zhou, Andrej Risteski
2021	J	`jnl`	An Online Learning Approach to Interpolation and Extrapolation in Domain Generalization. CoRR Elan Rosenfeld, Pradeep Ravikumar, Andrej Risteski
2021	J	`jnl`	Analyzing and Improving the Optimization Landscape of Noise-Contrastive Estimation. CoRR Bingbin Liu, Elan Rosenfeld, Pradeep Ravikumar, Andrej Risteski
2021	A	`conf`	Contrastive learning of strong-mixing continuous-time stochastic processes. AISTATS Bingbin Liu, Pradeep Ravikumar, Andrej Risteski
2021	J	`jnl`	Contrastive learning of strong-mixing continuous-time stochastic processes. CoRR Bingbin Liu, Pradeep Ravikumar, Andrej Risteski
2021	B	`conf`	Efficient sampling from the Bingham distribution. ALT Rong Ge, Holden Lee, Jianfeng Lu, Andrej Risteski
2021	J	`jnl`	Iterative Feature Matching: Toward Provable Domain Generalization with Logarithmic Environments. CoRR Yining Chen, Elan Rosenfeld, Mark Sellke, Tengyu Ma, Andrej Risteski
2021	A*	`conf`	Parametric Complexity Bounds for Approximating PDEs with Neural Networks. NeurIPS Tanya Marwah, Zachary C. Lipton, Andrej Risteski
2021	J	`jnl`	Parametric Complexity Bounds for Approximating PDEs with Neural Networks. CoRR Tanya Marwah, Zachary C. Lipton, Andrej Risteski
2021	A*	`conf`	Representational aspects of depth and conditioning in normalizing flows. ICML Frederic Koehler, Viraj Mehta, Andrej Risteski
2021	J	`jnl`	The Effects of Invertibility on the Representational Complexity of Encoders in Variational Autoencoders. CoRR Divyansh Pareek, Andrej Risteski
2021	—	`conf`	The Limitations of Limited Context for Constituency Parsing. ACL/IJCNLP (1) Yuchen Li, Andrej Risteski
2021	J	`jnl`	The Limitations of Limited Context for Constituency Parsing. CoRR Yuchen Li, Andrej Risteski
2021	A*	`conf`	The Risks of Invariant Risk Minimization. ICLR Elan Rosenfeld, Pradeep Kumar Ravikumar, Andrej Risteski
2021	A*	`conf`	Universal Approximation Using Well-Conditioned Normalizing Flows. NeurIPS Holden Lee, Chirag Pabbaraju, Anish Prasad Sevekari, Andrej Risteski
2021	J	`jnl`	Universal Approximation for Log-concave Distributions using Well-conditioned Normalizing Flows. CoRR Holden Lee, Chirag Pabbaraju, Anish Prasad Sevekari, Andrej Risteski
2021	J	`jnl`	Variational autoencoders in the presence of low-dimensional data: landscape and implicit bias. CoRR Frederic Koehler, Viraj Mehta, Andrej Risteski, Chenghui Zhou
2020	J	`jnl`	Efficient sampling from the Bingham distribution. CoRR Rong Ge, Holden Lee, Jianfeng Lu, Andrej Risteski
2020	A*	`conf`	Empirical Study of the Benefits of Overparameterization in Learning Latent Variable Models. ICML Rares-Darius Buhai, Yoni Halpern, Yoon Kim, Andrej Risteski, David A. Sontag
2020	J	`jnl`	Fast Convergence for Langevin Diffusion with Matrix Manifold Structure. CoRR Ankur Moitra, Andrej Risteski
2020	A*	`conf`	On Learning Language-Invariant Representations for Universal Machine Translation. ICML Han Zhao, Junjie Hu, Andrej Risteski
2020	J	`jnl`	On Learning Language-Invariant Representations for Universal Machine Translation. CoRR Han Zhao, Junjie Hu, Andrej Risteski
2020	J	`jnl`	Representational aspects of depth and conditioning in normalizing flows. CoRR Frederic Koehler, Viraj Mehta, Andrej Risteski
2020	J	`jnl`	The Risks of Invariant Risk Minimization. CoRR Elan Rosenfeld, Pradeep Ravikumar, Andrej Risteski
2019	—	`conf`	Approximability of Discriminators Implies Diversity in GANs. ICLR (Poster) Yu Bai, Tengyu Ma, Andrej Risteski
2019	J	`jnl`	Benefits of Overparameterization in Single-Layer Latent Variable Generative Models. CoRR Rares-Darius Buhai, Andrej Risteski, Yoni Halpern, David A. Sontag
2019	A*	`conf`	Mean-field approximation, convex hierarchies, and the optimality of correlation rounding: a unified perspective. STOC Vishesh Jain, Frederic Koehler, Andrej Risteski
2019	A*	`conf`	Sum-of-squares meets square loss: Fast rates for agnostic tensor completion. COLT Dylan J. Foster, Andrej Risteski
2019	J	`jnl`	Sum-of-squares meets square loss: Fast rates for agnostic tensor completion. CoRR Dylan J. Foster, Andrej Risteski
2019	—	`conf`	The Comparative Power of ReLU Networks and Polynomial Kernels in the Presence of Sparse Latent Structure. ICLR (Poster) Frederic Koehler, Andrej Risteski
2018	J	`jnl`	Approximability of Discriminators Implies Diversity in GANs. CoRR Yu Bai, Tengyu Ma, Andrej Risteski
2018	A*	`conf`	Beyond Log-concavity: Provable Guarantees for Sampling Multi-modal Distributions using Simulated Tempering Langevin Monte Carlo. NeurIPS Holden Lee, Andrej Risteski, Rong Ge
2018	—	`conf`	Do GANs learn the distribution? Some Theory and Empirics. ICLR (Poster) Sanjeev Arora, Andrej Risteski, Yi Zhang
2018	J	`jnl`	Linear Algebraic Structure of Word Senses, with Applications to Polysemy. Trans. Assoc. Comput. Linguistics Sanjeev Arora, Yuanzhi Li, Yingyu Liang, Tengyu Ma, Andrej Risteski
2018	J	`jnl`	Mean-field approximation, convex hierarchies, and the optimality of correlation rounding: a unified perspective. CoRR Vishesh Jain, Frederic Koehler, Andrej Risteski
2018	J	`jnl`	Representational Power of ReLU Networks and Polynomial Kernels: Beyond Worst-Case Analysis. CoRR Frederic Koehler, Andrej Risteski
2018	J	`jnl`	Simulated Tempering Langevin Monte Carlo II: An Improved Proof using Soft Markov Chain Decomposition. CoRR Rong Ge, Holden Lee, Andrej Risteski
2017	J	`jnl`	Beyond Log-concavity: Provable Guarantees for Sampling Multi-modal Distributions using Simulated Tempering Langevin Monte Carlo. CoRR Rong Ge, Holden Lee, Andrej Risteski
2017	J	`jnl`	Extending and Improving Wordnet via Unsupervised Word Embeddings. CoRR Mikhail Khodak, Andrej Risteski, Christiane Fellbaum, Sanjeev Arora
2017	—	`—`	New Techniques for Learning and Inference in Bayesian Models Andrej Risteski
2017	A*	`conf`	On the Ability of Neural Nets to Express Distributions. COLT Holden Lee, Rong Ge, Tengyu Ma, Andrej Risteski, Sanjeev Arora
2017	J	`jnl`	On the ability of neural nets to express distributions. CoRR Holden Lee, Rong Ge, Andrej Risteski, Tengyu Ma, Sanjeev Arora
2017	J	`jnl`	Provable benefits of representation learning. CoRR Sanjeev Arora, Andrej Risteski
2017	A*	`conf`	Provable learning of noisy-OR networks. STOC Sanjeev Arora, Rong Ge, Tengyu Ma, Andrej Risteski
2017	J	`jnl`	Theoretical limitations of Encoder-Decoder GAN architectures. CoRR Sanjeev Arora, Andrej Risteski, Yi Zhang
2016	J	`jnl`	A Latent Variable Model Approach to PMI-based Word Embeddings. Trans. Assoc. Comput. Linguistics Sanjeev Arora, Yuanzhi Li, Yingyu Liang, Tengyu Ma, Andrej Risteski
2016	—	`conf`	Algorithms and matching lower bounds for approximately-convex optimization. NIPS Andrej Risteski, Yuanzhi Li
2016	—	`conf`	Approximate maximum entropy principles via Goemans-Williamson with applications to provable variational methods. NIPS Andrej Risteski, Yuanzhi Li
2016	J	`jnl`	Approximate maximum entropy principles via Goemans-Williamson with applications to provable variational methods. CoRR Yuanzhi Li, Andrej Risteski
2016	A*	`conf`	How to calculate partition functions using convex programming hierarchies: provable bounds for variational methods. COLT Andrej Risteski
2016	J	`jnl`	How to calculate partition functions using convex programming hierarchies: provable bounds for variational methods. CoRR Andrej Risteski
2016	J	`jnl`	Linear Algebraic Structure of Word Senses, with Applications to Polysemy. CoRR Sanjeev Arora, Yuanzhi Li, Yingyu Liang, Tengyu Ma, Andrej Risteski
2016	—	`conf`	On Routing Disjoint Paths in Bounded Treewidth Graphs. SWAT Alina Ene, Matthias Mnich, Marcin Pilipczuk, Andrej Risteski
2016	J	`jnl`	Provable learning of Noisy-or Networks. CoRR Sanjeev Arora, Rong Ge, Tengyu Ma, Andrej Risteski
2016	—	`conf`	Recovery Guarantee of Non-negative Matrix Factorization via Alternating Updates. NIPS Yuanzhi Li, Yingyu Liang, Andrej Risteski
2016	J	`jnl`	Recovery Guarantee of Non-negative Matrix Factorization via Alternating Updates. CoRR Yuanzhi Li, Yingyu Liang, Andrej Risteski
2016	A*	`conf`	Recovery guarantee of weighted low-rank approximation via alternating minimization. ICML Yuanzhi Li, Yingyu Liang, Andrej Risteski
2016	J	`jnl`	Recovery guarantee of weighted low-rank approximation via alternating minimization. CoRR Yuanzhi Li, Yingyu Liang, Andrej Risteski
2015	A*	`conf`	Label optimal regret bounds for online local learning. COLT Pranjal Awasthi, Moses Charikar, Kevin A. Lai, Andrej Risteski
2015	J	`jnl`	Label optimal regret bounds for online local learning. CoRR Pranjal Awasthi, Moses Charikar, Kevin A. Lai, Andrej Risteski
2015	J	`jnl`	On Routing Disjoint Paths in Bounded Treewidth Graphs. CoRR Alina Ene, Matthias Mnich, Marcin Pilipczuk, Andrej Risteski
2015	—	`conf`	On some provably correct cases of variational inference for topic models. NIPS Pranjal Awasthi, Andrej Risteski
2015	J	`jnl`	On some provably correct cases of variational inference for topic models. CoRR Pranjal Awasthi, Andrej Risteski
2015	J	`jnl`	Random Walks on Context Spaces: Towards an Explanation of the Mysteries of Semantic Word Embeddings. CoRR Sanjeev Arora, Yuanzhi Li, Yingyu Liang, Tengyu Ma, Andrej Risteski
2014	J	`jnl`	Skeletal configurations of ribbon trees. Discret. Appl. Math. Howard Cheng, Satyan L. Devadoss, Brian Li, Andrej Risteski
2012	J	`jnl`	Skeletal Rigidity of Phylogenetic Trees CoRR Howard Cheng, Satyan L. Devadoss, Brian Li, Andrej Risteski
2012	—	`conf`	What makes a Tree a Straight Skeleton? CCCG Oswin Aichholzer, Howard Cheng, Satyan L. Devadoss, Thomas Hackl, Stefan Huber, Brian Li, Andrej Risteski