Gregory Valiant

151 papers A* 51A 10B 2Journal 78Unranked 8

Year	Rank	Type	Title / Venue / Authors
2026	—	`conf`	A Simple Geometric Proof of the Optimality of the Sequential Probability Ratio Test for Symmetric Bernoulli Hypotheses. SOSA Chirag Pabbaraju, Gregory Valiant, Rishi Verma
2025	A*	`conf`	A Generalized Trace Reconstruction Problem: Recovering a String of Probabilities. STOC Joey Rivkin, Gregory Valiant, Paul Valiant
2025	J	`jnl`	A Simple Geometric Proof of the Optimality of the Sequential Probability Ratio Test for Symmetric Bernoulli Hypotheses. CoRR Chirag Pabbaraju, Gregory Valiant, Rishi Verma
2025	A*	`conf`	Adaptive and Oblivious Statistical Adversaries Are Equivalent. STOC Guy Blanc, Gregory Valiant
2025	J	`jnl`	Attainability of Two-Point Testing Rates for Finite-Sample Location Estimation. CoRR Spencer Compton, Gregory Valiant
2025	J	`jnl`	On the Entropy Calibration of Language Models. CoRR Steven Cao, Gregory Valiant, Percy Liang
2025	J	`jnl`	Testing with Non-identically Distributed Samples. Trans. Mach. Learn. Res. Shivam Garg, Chirag Pabbaraju, Kirankumar Shiragur, Gregory Valiant
2024	J	`jnl`	A Generalized Trace Reconstruction Problem: Recovering a String of Probabilities. CoRR Joey Rivkin, Gregory Valiant, Paul Valiant
2024	J	`jnl`	Adaptive and oblivious statistical adversaries are equivalent. CoRR Guy Blanc, Gregory Valiant
2024	J	`jnl`	Discovering Data Structures: Nearest Neighbor Search and Beyond. CoRR Omar Salemohamed, Laurent Charlin, Shivam Garg, Vatsal Sharan, Gregory Valiant
2024	J	`jnl`	Efficient Convex Optimization Requires Superlinear Memory. J. ACM Annie Marsden, Vatsal Sharan, Aaron Sidford, Gregory Valiant
2024	A	`conf`	Matrix Multiplication in Quadratic Time and Energy? Towards a Fine-Grained Energy-Centric Church-Turing Thesis. ITCS Gregory Valiant
2024	A*	`conf`	Near-Optimal Mean Estimation with Unknown, Heteroskedastic Variances. STOC Spencer Compton, Gregory Valiant
2023	A*	`conf`	Efficient Convex Optimization Requires Superlinear Memory (Extended Abstract). IJCAI Annie Marsden, Vatsal Sharan, Aaron Sidford, Gregory Valiant
2023	A*	`conf`	Lexinvariant Language Models. NeurIPS Qian Huang, Eric Zelikman, Sarah Li Chen, Yuhuai Wu, Gregory Valiant, Percy Liang
2023	J	`jnl`	Lexinvariant Language Models. CoRR Qian Huang, Eric Zelikman, Sarah Li Chen, Yuhuai Wu, Gregory Valiant, Percy Liang
2023	J	`jnl`	Matrix Multiplication in Quadratic Time and Energy? Towards a Fine-Grained Energy-Centric Church-Turing Thesis. CoRR Gregory Valiant
2023	J	`jnl`	Near-Optimal Mean Estimation with Unknown, Heteroskedastic Variances. CoRR Spencer Compton, Gregory Valiant
2023	A*	`conf`	One-sided Matrix Completion from Two Observations Per Row. ICML Steven Cao, Percy Liang, Gregory Valiant
2023	J	`jnl`	One-sided Matrix Completion from Two Observations Per Row. CoRR Steven Cao, Percy Liang, Gregory Valiant
2023	A	`conf`	Online Pen Testing. ITCS Mingda Qiao, Gregory Valiant
2023	J	`jnl`	Testing with Non-identically Distributed Samples. CoRR Shivam Garg, Chirag Pabbaraju, Kirankumar Shiragur, Gregory Valiant
2022	A*	`conf`	Big-Step-Little-Step: Efficient Gradient Methods for Objectives with Multiple Scales. COLT Jonathan A. Kelner, Annie Marsden, Vatsal Sharan, Aaron Sidford, Gregory Valiant, Honglin Yuan
2022	A*	`conf`	Efficient Convex Optimization Requires Superlinear Memory. COLT Annie Marsden, Vatsal Sharan, Aaron Sidford, Gregory Valiant
2022	J	`jnl`	Efficient Convex Optimization Requires Superlinear Memory. CoRR Annie Marsden, Vatsal Sharan, Aaron Sidford, Gregory Valiant
2022	J	`jnl`	From Sand to Flour: The Next Leap in Granular Computing with NanoSort. CoRR Theo Jepsen, Stephen Ibanez, Gregory Valiant, Nick McKeown
2022	J	`jnl`	On the Statistical Complexity of Sample Amplification. CoRR Brian Axelrod, Shivam Garg, Yanjun Han, Vatsal Sharan, Gregory Valiant
2022	J	`jnl`	Online Pen Testing. CoRR Mingda Qiao, Gregory Valiant
2022	A*	`conf`	What Can Transformers Learn In-Context? A Case Study of Simple Function Classes. NeurIPS Shivam Garg, Dimitris Tsipras, Percy Liang, Gregory Valiant
2022	J	`jnl`	What Can Transformers Learn In-Context? A Case Study of Simple Function Classes. CoRR Shivam Garg, Dimitris Tsipras, Percy Liang, Gregory Valiant
2021	—	`conf`	Beyond Laurel/Yanny: An Autoencoder-Enabled Search for Polyperceivable Audio. ACL/IJCNLP (2) Kartik Chandra, Chuma Kabaghe, Gregory Valiant
2021	J	`jnl`	Big-Step-Little-Step: Efficient Gradient Methods for Objectives with Multiple Scales. CoRR Jonathan A. Kelner, Annie Marsden, Vatsal Sharan, Aaron Sidford, Gregory Valiant, Honglin Yuan
2021	A*	`conf`	Exponential Weights Algorithms for Selective Learning. COLT Mingda Qiao, Gregory Valiant
2021	J	`jnl`	Exponential Weights Algorithms for Selective Learning. CoRR Mingda Qiao, Gregory Valiant
2021	A	`conf`	Misspecification in Prediction Problems and Robustness via Improper Learning. AISTATS Annie Marsden, John C. Duchi, Gregory Valiant
2021	J	`jnl`	On Misspecification in Prediction Problems and Robustness via Improper Learning. CoRR Annie Marsden, John C. Duchi, Gregory Valiant
2021	A*	`conf`	Sinkhorn Label Allocation: Semi-Supervised Classification via Annealed Self-Training. ICML Kai Sheng Tai, Peter Bailis, Gregory Valiant
2021	J	`jnl`	Sinkhorn Label Allocation: Semi-Supervised Classification via Annealed Self-Training. CoRR Kai Sheng Tai, Peter Bailis, Gregory Valiant
2021	A*	`conf`	Stronger calibration lower bounds via sidestepping. STOC Mingda Qiao, Gregory Valiant
2020	A*	`conf`	Implicit regularization for deep neural networks driven by an Ornstein-Uhlenbeck like process. COLT Guy Blanc, Neha Gupta, Gregory Valiant, Paul Valiant
2020	—	`ch.`	Instance Optimal Distribution Testing and Learning. Beyond the Worst-Case Analysis of Algorithms Gregory Valiant, Paul Valiant
2020	A*	`conf`	On the Generalization Effects of Linear Transformations in Data Augmentation. ICML Sen Wu, Hongyang R. Zhang, Gregory Valiant, Christopher Ré
2020	J	`jnl`	On the Generalization Effects of Linear Transformations in Data Augmentation. CoRR Sen Wu, Hongyang R. Zhang, Gregory Valiant, Christopher Ré
2020	A*	`conf`	Sample Amplification: Increasing Dataset Size even when Learning is Impossible. ICML Brian Axelrod, Shivam Garg, Vatsal Sharan, Gregory Valiant
2020	J	`jnl`	Stronger Calibration Lower Bounds via Sidestepping. CoRR Mingda Qiao, Gregory Valiant
2020	A	`conf`	Sublinear Optimal Policy Value Estimation in Contextual Bandits. AISTATS Weihao Kong, Emma Brunskill, Gregory Valiant
2020	A*	`conf`	Worst-Case Analysis for Randomly Collected Data. NeurIPS Justin Y. Chen, Gregory Valiant, Paul Valiant
2019	A*	`conf`	A Polynomial Time Algorithm for Log-Concave Maximum Likelihood via Locally Exponential Families. NeurIPS Brian Axelrod, Ilias Diakonikolas, Alistair Stewart, Anastasios Sidiropoulos, Gregory Valiant
2019	J	`jnl`	A Polynomial Time Algorithm for Log-Concave Maximum Likelihood via Locally Exponential Families. CoRR Brian Axelrod, Ilias Diakonikolas, Anastasios Sidiropoulos, Alistair Stewart, Gregory Valiant
2019	B	`conf`	A Surprising Density of Illusionable Natural Speech. CogSci Melody Y. Guan, Gregory Valiant
2019	J	`jnl`	A Surprising Density of Illusionable Natural Speech. CoRR Melody Y. Guan, Gregory Valiant
2019	A*	`conf`	A Theory of Selective Prediction. COLT Mingda Qiao, Gregory Valiant
2019	J	`jnl`	A Theory of Selective Prediction. CoRR Mingda Qiao, Gregory Valiant
2019	A*	`conf`	Compressed Factorization: Fast and Accurate Low-Rank Factorization of Compressively-Sensed Data. ICML Vatsal Sharan, Kai Sheng Tai, Peter Bailis, Gregory Valiant
2019	A*	`conf`	Equivariant Transformer Networks. ICML Kai Sheng Tai, Peter Bailis, Gregory Valiant
2019	J	`jnl`	Equivariant Transformer Networks. CoRR Kai Sheng Tai, Peter Bailis, Gregory Valiant
2019	J	`jnl`	How bad is worst-case data if you know where it comes from? CoRR Justin Y. Chen, Gregory Valiant, Paul Valiant
2019	J	`jnl`	Implicit regularization for deep neural networks driven by an Ornstein-Uhlenbeck like process. CoRR Guy Blanc, Neha Gupta, Gregory Valiant, Paul Valiant
2019	A*	`conf`	Making AI Forget You: Data Deletion in Machine Learning. NeurIPS Antonio Ginart, Melody Y. Guan, Gregory Valiant, James Zou
2019	J	`jnl`	Making AI Forget You: Data Deletion in Machine Learning. CoRR Antonio Ginart, Melody Y. Guan, Gregory Valiant, James Zou
2019	A*	`conf`	Maximum Likelihood Estimation for Learning Populations of Parameters. ICML Ramya Korlakai Vinayak, Weihao Kong, Gregory Valiant, Sham M. Kakade
2019	J	`jnl`	Maximum Likelihood Estimation for Learning Populations of Parameters. CoRR Ramya Korlakai Vinayak, Weihao Kong, Gregory Valiant, Sham M. Kakade
2019	J	`jnl`	Memory-Sample Tradeoffs for Linear Regression with Small Error. CoRR Vatsal Sharan, Aaron Sidford, Gregory Valiant
2019	A*	`conf`	Memory-sample tradeoffs for linear regression with small error. STOC Vatsal Sharan, Aaron Sidford, Gregory Valiant
2019	J	`jnl`	Sample Amplification: Increasing Dataset Size even when Learning is Impossible. CoRR Brian Axelrod, Shivam Garg, Vatsal Sharan, Gregory Valiant
2019	J	`jnl`	Sublinear Optimal Policy Value Estimation in Contextual Bandits. CoRR Weihao Kong, Gregory Valiant, Emma Brunskill
2018	A*	`conf`	A Data Prism: Semi-verified learning in the small-alpha regime. COLT Michela Meister, Gregory Valiant
2018	A*	`conf`	A Spectral View of Adversarially Robust Features. NeurIPS Shivam Garg, Vatsal Sharan, Brian Hu Zhang, Gregory Valiant
2018	J	`jnl`	A Spectral View of Adversarially Robust Features. CoRR Shivam Garg, Vatsal Sharan, Brian Hu Zhang, Gregory Valiant
2018	J	`jnl`	An Efficient Algorithm for High-Dimensional Log-Concave Maximum Likelihood. CoRR Brian Axelrod, Gregory Valiant
2018	A*	`conf`	Approximating the Spectrum of a Graph. KDD David Cohen-Steiner, Weihao Kong, Christian Sohler, Gregory Valiant
2018	A*	`conf`	Estimating Learnability in the Sublinear Data Regime. NeurIPS Weihao Kong, Gregory Valiant
2018	J	`jnl`	Estimating Learnability in the Sublinear Data Regime. CoRR Weihao Kong, Gregory Valiant
2018	A	`conf`	Learning Discrete Distributions from Untrusted Batches. ITCS Mingda Qiao, Gregory Valiant
2018	A*	`conf`	Prediction with a short memory. STOC Vatsal Sharan, Sham M. Kakade, Percy Liang, Gregory Valiant
2018	A	`conf`	Recovering Structured Probability Matrices. ITCS Qingqing Huang, Sham M. Kakade, Weihao Kong, Gregory Valiant
2018	A	`conf`	Resilience: A Criterion for Learning in the Presence of Arbitrary Outliers. ITCS Jacob Steinhardt, Moses Charikar, Gregory Valiant
2018	—	`conf`	Sketching Linear Classifiers over Data Streams. SIGMOD Conference Kai Sheng Tai, Vatsal Sharan, Peter Bailis, Gregory Valiant
2017	J	`jnl`	A Data Prism: Semi-Verified Learning in the Small-Alpha Regime. CoRR Michela Meister, Gregory Valiant
2017	J	`jnl`	An Automatic Inequality Prover and Instance Optimal Identity Testing. SIAM J. Comput. Gregory Valiant, Paul Valiant
2017	J	`jnl`	Approximating the Spectrum of a Graph. CoRR David Cohen-Steiner, Weihao Kong, Christian Sohler, Gregory Valiant
2017	J	`jnl`	Estimating the Unseen: Improved Estimators for Entropy and Other Properties. J. ACM Gregory Valiant, Paul Valiant
2017	A*	`conf`	Estimating the unseen from multiple populations. ICML Aditi Raghunathan, Gregory Valiant, James Zou
2017	J	`jnl`	Estimating the unseen from multiple populations. CoRR Aditi Raghunathan, Gregory Valiant, James Zou
2017	J	`jnl`	Finding Heavily-Weighted Features in Data Streams. CoRR Kai Sheng Tai, Vatsal Sharan, Peter Bailis, Gregory Valiant
2017	J	`jnl`	Learning Discrete Distributions from Untrusted Batches. CoRR Mingda Qiao, Gregory Valiant
2017	—	`conf`	Learning Overcomplete HMMs. NIPS Vatsal Sharan, Sham M. Kakade, Percy Liang, Gregory Valiant
2017	J	`jnl`	Learning Overcomplete HMMs. CoRR Vatsal Sharan, Sham M. Kakade, Percy Liang, Gregory Valiant
2017	—	`conf`	Learning Populations of Parameters. NIPS Kevin Tian, Weihao Kong, Gregory Valiant
2017	A*	`conf`	Learning from untrusted data. STOC Moses Charikar, Jacob Steinhardt, Gregory Valiant
2017	J	`jnl`	Optimally Learning Populations of Parameters. CoRR Kevin Tian, Weihao Kong, Gregory Valiant
2017	A*	`conf`	Orthogonalized ALS: A Theoretically Principled Tensor Decomposition Algorithm for Practical Use. ICML Vatsal Sharan, Gregory Valiant
2017	J	`jnl`	Orthogonalized ALS: A Theoretically Principled Tensor Decomposition Algorithm for Practical Use. CoRR Vatsal Sharan, Gregory Valiant
2017	J	`jnl`	Resilience: A Criterion for Learning in the Presence of Arbitrary Outliers. CoRR Jacob Steinhardt, Moses Charikar, Gregory Valiant
2017	J	`jnl`	There and Back Again: A General Approach to Learning Sparse Models. CoRR Vatsal Sharan, Kai Sheng Tai, Peter Bailis, Gregory Valiant
2016	—	`conf`	Avoiding Imposters and Delinquents: Adversarial Crowdsourcing and Peer Prediction. NIPS Jacob Steinhardt, Gregory Valiant, Moses Charikar
2016	J	`jnl`	Avoiding Imposters and Delinquents: Adversarial Crowdsourcing and Peer Prediction. CoRR Jacob Steinhardt, Gregory Valiant, Moses Charikar
2016	J	`jnl`	Information Theoretically Secure Databases. CoRR Gregory Valiant, Paul Valiant
2016	J	`jnl`	Information Theoretically Secure Databases. Electron. Colloquium Comput. Complex. Gregory Valiant, Paul Valiant
2016	A*	`conf`	Instance optimal learning of discrete distributions. STOC Gregory Valiant, Paul Valiant
2016	J	`jnl`	Learning from Untrusted Data. CoRR Moses Charikar, Jacob Steinhardt, Gregory Valiant
2016	A*	`conf`	Memory, Communication, and Statistical Queries. COLT Jacob Steinhardt, Gregory Valiant, Stefan Wager
2016	J	`jnl`	Prediction with a Short Memory. CoRR Sham M. Kakade, Percy Liang, Vatsal Sharan, Gregory Valiant
2016	J	`jnl`	Recovering Structured Probability Matrices. CoRR Qingqing Huang, Sham M. Kakade, Weihao Kong, Gregory Valiant
2016	J	`jnl`	Spectrum Estimation from Samples. CoRR Weihao Kong, Gregory Valiant
2016	A	`conf`	Stochastic Streams: Sample Complexity vs. Space Complexity. ESA Michael S. Crouch, Andrew McGregor, Gregory Valiant, David P. Woodruff
2015	J	`jnl`	Finding Correlations in Subquadratic Time, with Applications to Learning Parities and the Closest Pair Problem. J. ACM Gregory Valiant
2015	J	`jnl`	Instance Optimal Learning. CoRR Gregory Valiant, Paul Valiant
2015	J	`jnl`	Memory, Communication, and Statistical Queries. Electron. Colloquium Comput. Complex. Jacob Steinhardt, Gregory Valiant, Stefan Wager
2015	—	`conf`	Testing Closeness With Unequal Sized Samples. NIPS Bhaswar B. Bhattacharya, Gregory Valiant
2015	J	`jnl`	Testing Closeness With Unequal Sized Samples. CoRR Bhaswar B. Bhattacharya, Gregory Valiant
2014	A*	`conf`	An Automatic Inequality Prover and Instance Optimal Identity Testing. FOCS Gregory Valiant, Paul Valiant
2014	A*	`conf`	Learning Polynomials with Neural Networks. ICML Alexandr Andoni, Rina Panigrahy, Gregory Valiant, Li Zhang
2014	A*	`conf`	Learning Sparse Polynomial Functions. SODA Alexandr Andoni, Rina Panigrahy, Gregory Valiant, Li Zhang
2014	A*	`conf`	Least Squares Revisited: Scalable Approaches for Multi-class Prediction. ICML Alekh Agarwal, Sham M. Kakade, Nikos Karampatziakis, Le Song, Gregory Valiant
2014	A*	`conf`	Optimal Algorithms for Testing Closeness of Discrete Distributions. SODA Siu On Chan, Ilias Diakonikolas, Paul Valiant, Gregory Valiant
2014	A*	`conf`	Satisfiability and Evolution. FOCS Adi Livnat, Christos H. Papadimitriou, Aviad Rubinstein, Gregory Valiant, Andrew Wan
2013	J	`jnl`	Computation in anonymous networks. CoRR Elchanan Mossel, Anupam Prakash, Gregory Valiant
2013	—	`conf`	Estimating the Unseen: Improved Estimators for Entropy and other Properties. NIPS Paul Valiant, Gregory Valiant
2013	J	`jnl`	Instance-by-instance optimal identity testing. Electron. Colloquium Comput. Complex. Gregory Valiant, Paul Valiant
2013	J	`jnl`	Least Squares Revisited: Scalable Approaches for Multi-class Prediction. CoRR Alekh Agarwal, Sham M. Kakade, Nikos Karampatziakis, Le Song, Gregory Valiant
2013	J	`jnl`	Optimal Algorithms for Testing Closeness of Discrete Distributions. CoRR Siu On Chan, Ilias Diakonikolas, Gregory Valiant, Paul Valiant
2013	A*	`conf`	Testing SODA Constantinos Daskalakis, Ilias Diakonikolas, Rocco A. Servedio, Gregory Valiant, Paul Valiant
2012	—	`—`	Algorithmic Approaches to Statistical Questions. Gregory Valiant
2012	J	`jnl`	Beating brute-force: Improved algorithms for finding correlations, and related problems. Tiny Trans. Comput. Sci. Gregory Valiant
2012	J	`jnl`	Disentangling Gaussians. Commun. ACM Adam Tauman Kalai, Ankur Moitra, Gregory Valiant
2012	J	`jnl`	Finding Correlations in Subquadratic Time, with Applications to Learning Parities and Juntas with Noise. Electron. Colloquium Comput. Complex. Gregory Valiant
2012	A*	`conf`	Finding Correlations in Subquadratic Time, with Applications to Learning Parities and Juntas. FOCS Gregory Valiant
2012	J	`jnl`	Size and Treewidth Bounds for Conjunctive Queries. J. ACM Georg Gottlob, Stephanie Tien Lee, Gregory Valiant, Paul Valiant
2011	A	`conf`	Best-Response Mechanisms. ICS Noam Nisan, Michael Schapira, Gregory Valiant, Aviv Zohar
2011	A*	`conf`	Best-response auctions. EC Noam Nisan, Michael Schapira, Gregory Valiant, Aviv Zohar
2011	A*	`conf`	Estimating the unseen: an n/log(n)-sample estimator for entropy and support size, shown optimal via new CLTs. STOC Gregory Valiant, Paul Valiant
2011	A*	`conf`	Incentive-compatible distributed greedy protocols. PODC Noam Nisan, Michael Schapira, Gregory Valiant, Aviv Zohar
2011	J	`jnl`	Testing $k$-Modal Distributions: Optimal Algorithms via Reductions CoRR Constantinos Daskalakis, Ilias Diakonikolas, Rocco A. Servedio, Gregory Valiant, Paul Valiant
2011	A*	`conf`	The Power of Linear Estimators. FOCS Gregory Valiant, Paul Valiant
2011	J	`jnl`	When is it best to best-respond? SIGecom Exch. Noam Nisan, Michael Schapira, Gregory Valiant, Aviv Zohar
2010	J	`jnl`	A CLT and tight lower bounds for estimating entropy. Electron. Colloquium Comput. Complex. Gregory Valiant, Paul Valiant
2010	A	`conf`	A New Look at Selfish Routing. ICS Christos H. Papadimitriou, Gregory Valiant
2010	J	`jnl`	Braess's Paradox in large random graphs. Random Struct. Algorithms Gregory Valiant, Tim Roughgarden
2010	J	`jnl`	Designing Network Protocols for Good Equilibria. SIAM J. Comput. Ho-Lin Chen, Tim Roughgarden, Gregory Valiant
2010	A*	`conf`	Efficiently learning mixtures of two Gaussians. STOC Adam Tauman Kalai, Ankur Moitra, Gregory Valiant
2010	J	`jnl`	Estimating the unseen: A sublinear-sample canonical estimator of distributions. Electron. Colloquium Comput. Complex. Gregory Valiant, Paul Valiant
2010	B	`conf`	On Learning Algorithms for Nash Equilibria. SAGT Constantinos Daskalakis, Rafael M. Frongillo, Christos H. Papadimitriou, George Pierrakos, Gregory Valiant
2010	J	`jnl`	Settling the Polynomial Learnability of Mixtures of Gaussians CoRR Ankur Moitra, Gregory Valiant
2010	A*	`conf`	Settling the Polynomial Learnability of Mixtures of Gaussians. FOCS Ankur Moitra, Gregory Valiant
2009	A*	`conf`	On the complexity of Nash equilibria of action-graph games. SODA Constantinos Daskalakis, Grant Schoenebeck, Gregory Valiant, Paul Valiant
2009	J	`jnl`	Size Bounds for Conjunctive Queries with General Functional Dependencies CoRR Gregory Valiant, Paul Valiant
2009	A*	`conf`	Size and treewidth bounds for conjunctive queries. PODS Georg Gottlob, Stephanie Tien Lee, Gregory Valiant
2008	A*	`conf`	Designing networks with good equilibria. SODA Ho-Lin Chen, Tim Roughgarden, Gregory Valiant
2008	J	`jnl`	On the Complexity of Nash Equilibria of Action-Graph Games CoRR Constantinos Daskalakis, Grant Schoenebeck, Gregory Valiant, Paul Valiant
2006	A*	`conf`	Braess's paradox in large random graphs. EC Gregory Valiant, Tim Roughgarden