Introduces GANGs to model GANs as games, solving them with RB-NE.
problem Training difficulties in GANs.
method Explicitly models GANs as a zero-sum game between generator and classifier using mixed strategies, defines resource-bounded best responses and Nash Equilibrium.
result Solves GANGs with RB-NE, demonstrating better performance than standard GANs.
This paper improves GANs by ensuring local equilibria are also Nash equilibria, leading to better solutions.
problem Current GAN training methods often converge to local Nash equilibria, which may not be optimal.
method Formalizes GANs as finite games in mixed strategies, ensuring every local equilibrium is a Nash equilibrium.
result The proposed method converges to a resource-bounded Nash equilibrium, producing solutions closer to theoretical predictions.
Transformers use ReLUs to approximate softmax efficiently.
problem Analyzing resource usage in softmax transformer models.
method Translating ReLU approximation results to softmax attention mechanisms.
result Economic resource bounds for softmax attention mechanisms.
The study explains how humans make sub-optimal decisions using learned approximations.
problem Human decision-making often deviates from optimal strategies.
method Combining meta-learning models with resource-bounded objectives.
result Empirical evidence shows patterns in human decision-making that align with the model.
Self-Tuning Networks optimize hyperparameters using bilevel optimization and gated best-response functions.
problem Optimizing hyperparameters for neural networks.
method Bilevel optimization with gated best-response functions to adapt regularization hyperparameters online.
result Self-Tuning Networks outperform fixed hyperparameter values on large-scale deep learning problems.
The aim of this work is to address the question of whether we can in principle design rational decision-making agents or artificial intelligences embedded in computable physics such that their decisions are optimal in reasonable mathematical senses. Recent developments in rare event probability estimation, recursive ba…
Study best-response learning dynamics in zero-sum polymatrix games under full and minimal information settings.
problem Learning dynamics in zero-sum polymatrix games under different information settings.
method Two-timescale learning dynamics combining smoothed best-response updates and TD-learning for estimating local payoff functions.
result Polynomial-time finite-sample guarantees for convergence to an ε-Nash equilibrium in the minimal information case.
Neural operators approximate Stackelberg game solutions.
problem Intractability of follower's best-response operator in dynamic Stackelberg games.
method Used attention-based neural operators to approximate the best-response operator.
result Approximate best-response operator yields close game value.
Two methods for model adaptation compared; fine-tuning outperforms Best-of-N in realizable settings.
problem Comparing methods for adapting large language models to new tasks.
method Supervised fine-tuning vs. Best-of-N approach.
result Supervised fine-tuning outperforms Best-of-N in realizable settings.
New method for sparse kernel selection improves prediction accuracy.
problem Sparse Multiple Kernel Learning for binary classification.
method Alternating best response algorithm with semidefinite relaxations.
result Method outperforms state-of-the-art MKL approaches in prediction accuracy.
Improved hypernetwork for efficient neural network hyperparameter tuning.
problem Efficiently optimizing hyperparameters in neural networks.
method Proposed Δ-STN architecture focusing on accurate best-response Jacobian approximation. result Significantly improved hyperparameter tuning accuracy and stability.
Paper introduces metrics for evaluating multi-agent policies using best response dynamics.
problem Evaluation and ranking of multi-agent policies in reinforcement learning.
method Adopting strict best response dynamics (SBRD) to model selfish behaviors, proposing perturbed SBRD for dynamic and non-stationary settings.
result Proposed perturbed SBRD can observe policies with maximum metrics and differ from optimal by any given tolerance.
New algorithms achieve decision calibration without sample complexity dependent on feature dimension.
problem Achieving decision calibration for nonlinear loss functions with polynomial sample complexity.
method Developed smooth relaxation of decision calibration, enabling dimension-free algorithms.
result Efficient algorithms post-process predictors to satisfy decision calibration without worsening accuracy.
Best-of-N sampling reveals reward targets from preference data, influencing N and base distribution choices.
problem Understanding reward extraction from Best-of-N preference data and optimal N and base distribution choices.
method Specialized analysis of preference data via induced conditional distribution, deriving reward targets and design principles.
result Reward targets are explicit functions of N and base distribution, and bounded-class minimizers approach these targets as N grows.
Paper studies zero-sum games with noisy observations and identifies equilibrium conditions.
problem Zero-sum games with noisy observations of the leader's actions.
method Analyzes the equilibrium of games with noisy action observability, identifies necessary conditions for uniqueness, and investigates the cardinality of best responses.
result The noisy observations significantly impact the cardinality of the follower's set of best responses, and under certain conditions, this set becomes a singleton almost surely.
ISMCTS-BR learns best responses in large games, approximating worst-case performance.
problem Learning robustness to worst-case outcomes in large games.
method ISMCTS-BR, a scalable search-based algorithm for deep reinforcement learning.
result ISMCTS-BR approximates worst-case performance in large games.
Best-of-Majority improves inference performance in Pass@k settings.
problem Inference in difficult tasks often underperforms with single-shot selection methods.
method Combining majority voting and Best-of-N, Best-of-Majority restricts candidates to high-frequency responses.
result Best-of-Majority achieves minimax optimal regret and outperforms other methods.
Deep RL improves Diplomacy performance, outperforming previous methods.
problem Applying RL to complex, many-agent, simultaneous-move games like Diplomacy.
method Best response policy iteration and fictitious play approximation.
result Deep RL agents convincingly outperform previous Diplomacy agents.
New algorithm improves inference-time alignment without reward hacking.
problem Improving quality of responses from language models with limited compute.
method Inference-time alignment, focusing on extttInferenceTimePessimism algorithm. result Optimal performance and scaling-monotonicity of extttInferenceTimePessimism. Paper improves emotion expression in AI chatbots.
problem AI chatbots generate responses that lack emotion.
method Developed neural models to express specific emotions in generated responses.
result An encoder-decoder model with multiple attention layers performs best in expressing required emotion.
Adversarial learning improves multi-turn dialogue response generation.
problem Generating coherent and diverse multi-turn dialogue responses.
method hredGAN based on conditional GANs with HRED generator and word-level RNN discriminator.
result hredGAN generates more informative and diverse responses with limited data.
Study of portfolio management under relative performance concerns using mean field games.
problem Portfolio management problems under relative performance concerns.
method Forward utilities of CARA type, mean field games, best response and equilibrium strategies.
result Solve forward-utility finite player game and mean-field game under asset specialization.
Enhances preference learning by incorporating response times into binary choices.
problem Limited information from binary choices about preference strength.
method Combines choices and response times using the EZ diffusion model.
result Response times improve utility estimation for strong preferences.
Parallel algorithm finds sparse solutions for nonconvex problems.
problem Nonconvex sparsity-regularized rank minimization.
method Parallel best-response algorithm with exact line search.
result Guaranteed convergence to a stationary point.
NeuPL learns diverse policies in strategy games efficiently.
problem Iterative training of policies in strategy games leads to under-trained good-responses and wasteful repetition.
method NeuPL uses a single conditional model to represent a population of policies, offering convergence guarantees and transfer learning.
result NeuPL achieves better performance and efficiency across various domains, enabling access to novel strategies.
We consider learning high-dimensional multi-response linear models with structured parameters. By exploiting the noise correlations among responses, we propose an alternating estimation (AltEst) procedure to estimate the model parameters based on the generalized Dantzig selector. Under suitable sample size and resampli…
New algorithms achieve logarithmic regret in KL-regularized Markov games.
problem Improving sample efficiency in game-theoretic settings with KL regularization.
method Developed OMG and SOMG algorithms for matrix and Markov games, using best response sampling and superoptimistic bonuses.
result Logarithmic regret in T that scales inversely with KL regularization strength β. Flexible models predict US Census survey response rates.
problem Predicting survey response rates in the US Census Bureau.
method Nonparametric additive models with structured interactions using ℓ0-based penalization.
result Models lead to predictions comparable to black-box methods but are interpretable.
New method optimizes language model performance for test-time strategies.
problem Mismatch between training objectives and test-time deployment of large language models.
method Tail-Extrapolated estimators to approximate best-of-N performance from limited training rollouts.
result Improved performance of best-of-N deployment across various models and datasets.
Algorithm solves online binary classification and infinite games using ERM oracle.
problem Online learning and solving infinite games with computationally inefficient oracles.
method Proposes an algorithm relying solely on ERM oracle calls for online binary classification and nonparametric games.
result Achieves finite and sublinearly growing regret in various settings.
Model shows how advisors can manipulate naive investors.
problem How financial advisors manipulate naive investors.
method Agent-Based Model with Nash equilibria and best response functions.
result Greediness/naivety of investors emerge naturally from the model.
When simulating a complex stochastic system, the behavior of output response depends on input parameters estimated from finite real-world data, and the finiteness of data brings input uncertainty into the system. The quantification of the impact of input uncertainty on output response has been extensively studied. Most…
New algorithm for solving minimax problems over distributions converges to Nash equilibrium.
problem Solving minimax problems over probability distributions.
method Symmetric Mean-field Langevin Dynamics (MFL-AG and MFL-ABR) with weighted averaging and best response dynamics.
result Converges to mixed Nash equilibrium with average-iterate and last-iterate convergence.
Digital personas improve survey results for stable attributes but fail for subjective responses.
problem When can digital personas reliably approximate human survey findings?
method Using LISS panel, constructed personas from background variables and survey histories, tested against held-out post-cutoff answers.
result Digital personas improve alignment with human response distributions for stable attributes but fail for subjective responses.
Bayesian model ranks treatments in multi-response experiments.
problem Identifying the best treatment among competing ideal properties.
method Bayesian approach with Markov Chain Monte Carlo algorithm.
result Reliable inference of treatment ranks in practice.
New algorithms solve games to find optimal exploration strategies.
problem Sequentially gather information to answer queries about stochastic environments.
method Interpret optimisation problem as a game and use iterative strategies to estimate its saddle point.
result First finite confidence guarantees for pure exploration in exponential families.
No best k-layer neural network approximations exist in general for common activations.
problem The non-existence of best k-layer neural network approximations for common activation functions.
method Empirical Risk Minimization (ERM) for neural networks with k layers and various activation functions.
result For k = 2 and common activations like ReLU, hyperbolic tangent, and sigmoid, the ERM infimum is not attainable in general.
Deep Claim predicts payer responses from claims data using deep learning.
problem Predicting payer responses from claims data to improve healthcare performance.
method Learning complex dependencies in claim inputs to create a compact representation, then using deep learning to predict responses.
result Deep Claim improves claim denial prediction by 22.21%.
Paper presents IPVI for unbiased DGP inference with efficiency.
problem Intractable exact inference in deep Gaussian processes.
method Implicit posterior variational inference framework.
result IPVI achieves unbiased posterior belief with efficiency.
A fast algorithm selects best subsets in high-dimensional models.
problem Identifying sparse models in high-dimensional generalized linear models.
method Splicing technique for fast and consistent best subset selection.
result Our algorithm achieves high certainty in selecting best subsets with polynomial computational complexity.
Heterogeneous SVO leads to diverse policies in sequential social dilemmas.
problem Understanding how diverse social value orientations affect behavior in sequential social dilemmas.
method Extending prior reinforcement learning studies, we instantiated heterogeneous SVO in a sequential social dilemma setting and measured task-specific diversity metrics.
result Heterogeneous SVO leads to meaningfully diverse policies across various incentive structures.
The paper presents a machine learning approach to multidimensional item response theory.
problem Modeling and predicting student performance from assessment data.
method Inspired by collaborative filtering, the paper defines a general class of models using penalized joint maximum likelihood (JML) for estimation and cross-validation for model selection.
result The high-dimensional model fit to large and sparse data does not lend itself well to traditional factor interpretation.
ONN learns different representations for different operations to improve user response prediction.
problem Improving user response prediction in online advertising and recommendation systems.
method Proposes Operation-aware Neural Networks (ONN) to learn different representations for different operations.
result ONN consistently outperforms state-of-the-art models in both offline and online environments.
DLGNet improves dialogue response generation by leveraging transformer architecture.
problem Lack of relevance, diversity, and coherence in dialogue responses.
method Transformer-based model for dialogue response generation, incorporating long-range structures and random paddings.
result Significant improvements over state-of-the-art models on multiple datasets.
New RL algorithms find SNE in Markov games with myopic followers.
problem Finding SNE in Markov games with myopic followers.
method Optimistic and pessimistic variants of least-squares value iteration, incorporating function approximation.
result First provably efficient RL algorithms for SNEs in general-sum Markov games with myopic followers.
Optimal ability estimation in adaptive testing with binary responses.
problem Estimating a continuous ability parameter from sequential binary responses.
method Adaptive selection of questions to maximize Fisher information, updating estimate using method-of-moments, and deciding accuracy with a test statistic.
result Fisher-tracking strategy achieves optimal performance in fixed-confidence and fixed-budget regimes.
New method improves robustness in partially observable domains by training against latent distribution shifts.
problem Challenges in robustness under latent distribution shift in partially observable reinforcement learning.
method Formalizes adversarial latent-initial-state POMDP, proves minimax principle, derives best-response inequalities.
result Reduces robustness gaps from 10.3 to 3.1 shots with targeted exposure to shifted latent distributions.
Proposes DeepSDRF for continuous treatment recommendation from clinical survival data.
problem Continuous treatment recommendation in medical settings with survival data.
method Deep Survival Dose Response Function (DeepSDRF) for learning conditional average dose response (CADR) function.
result Similar performance of recommender algorithms based on random search and reinforcement learning.