This work proves win rate is key to understanding preference learning.
problem Understanding preference learning from generative models.
method Analyzing preference learning methods as win rate optimization or non-WRO.
result Proves win rate is the only evaluation respecting preferences and prevalences.
Bayesian optimization improved AlphaGo's win-rate from 50% to 66.5%.
problem Hyper-parameter tuning for machine learning models.
method Bayesian optimization for hyper-parameter tuning.
result Bayesian optimization improved AlphaGo's performance in self-play games.
Faster WIND accelerates iterative BOND for LLM alignment.
problem Iterative BOND is inefficient in practice due to sample and computation inefficiency.
method Unified game-theoretic connection to self-play alignment, WIND framework with efficient algorithms.
result WIND variant achieves superior sample efficiency and faster computation.
New bid shading algorithm reduces costs by 55%.
problem Minimizing costs in online advertising auctions.
method Win-rate prediction and surplus maximization using logistic regression.
result 7% more profit for advertisers compared to benchmarks.
Improved MACD trading strategies with other indicators for better performance.
problem Evaluating the effectiveness of MACD-based trading strategies in the US stock market.
method Backtested various MACD-based trading strategies on US stock indices using Python.
result Win-rate of MACD strategies improved with other momentum indicators, leading to a new VPVMA indicator.
Paper presents TG and TL for efficient RL in StarCraft II.
problem Efficient reinforcement learning for StarCraft II.
method Abstract forward model (TG) combined with transfer learning (TL).
result 99% win-rate on 64x64 map with 1.08 hours of training.
DR-MCTS improves decision quality and sample efficiency in complex environments.
problem Improving decision quality and sample efficiency in complex environments.
method Integrates Doubly Robust off-policy estimation into Monte Carlo Tree Search (MCTS).
result DR-MCTS achieves superior performance in Tic-Tac-Toe and VirtualHome tasks.
StarCraft II poses a grand challenge for reinforcement learning. The main difficulties of it include huge state and action space and a long-time horizon. In this paper, we investigate a hierarchical reinforcement learning approach for StarCraft II. The hierarchy involves two levels of abstraction. One is the macro-acti…
TabPFN-2.5 boosts tabular AI performance, especially for large datasets.
problem Improving tabular AI models for large datasets and complex feature sets.
method Advanced tabular foundation model with data augmentation and distillation techniques.
result TabPFN-2.5 outperforms tuned tree-based models and AutoGluon 1.4 on benchmark datasets.
Model selection for time series forecasting can be biased by the distribution of scores.
problem Model selection for probabilistic forecasting on time series data.
method Using proper scoring rules to aggregate scores across multiple time series.
result The mean score is immune to the skewness of the score distribution.
SPPO optimizes language model alignment by treating preferences as a game and achieving state-of-the-art performance.
problem Capturing intransitivity and irrationality in human preferences for accurate language model alignment.
method Self-play-based approach to identify Nash equilibrium policy through iterative policy updates.
result SPPO achieves state-of-the-art win-rate of 28.53% on AlpacaEval 2.0 without external supervision.
Adversarial policies beat superhuman Go AI systems.
problem Vulnerability of superhuman AI systems to adversarial attacks.
method Training adversarial policies to trick KataGo into making blunders.
result Adversarial policies achieve >97% win rate against KataGo at superhuman settings.
Study improves StarCraft bot's strategy selection with partial observations.
problem Selecting effective strategies in real-time strategy games with limited information.
method Utilized full game state information during training to predict opponent strategies.
result Substantial win rate improvements over a fixed-strategy baseline.
We explore a new way to evaluate generative models using insights from evaluation of competitive games between human players. We show experimentally that tournaments between generators and discriminators provide an effective way to evaluate generative models. We introduce two methods for summarizing tournament outcomes…
We rebias estimates to improve interval calibration and prediction accuracy.
problem Constructing accurate intervals for noisy and biased estimates.
method Empirical Bayes rebiasing strategy that learns bias distribution from data.
result Substantial precision gains in prediction-powered inference.
This paper introduces a gradient analysis framework to improve language model performance by rewarding good examples and penalizing bad ones.
problem Improving language model output quality by penalizing bad examples.
method Gradient analysis of loss functions to reward good examples and penalize bad ones.
result ExMATE is superior to MLE and combining DPO with ExMATE enhances performance.
Gamblers lose in long bets despite casino claims, study shows.
problem Gamblers losing in seemingly fair gambling scenarios.
method Theoretical and experimental analysis of two-armed Futurity slot machines.
result Casinos have a win rate greater than 50% in long bets, making them unfair.
MallowsPO enhances LLM fine-tuning with a dispersion index of human preferences.
problem Lack of diversity in human preferences in DPO.
method Developed a dispersion index based on Mallows' theory to characterize preference diversity.
result Demonstrated improved performance in various tasks using the dispersion index.
SharedRep-RLHF learns shared traits for diverse groups, improving fairness and performance.
problem Uniform-reward RLHF fails to capture diverse preferences, leading to unfairness.
method SharedRep-RLHF learns shared traits among various groups, improving fairness and performance.
result SharedRep-RLHF outperforms MaxMin-RLHF by up to 20% in win rate.
Generative framework improves causal estimation from observational data.
problem Estimating individualized treatment effects from non-randomized data.
method Importance-Weighted Diffusion Distillation (IWDD) combining diffusion models and IPW.
result IWDD achieves state-of-the-art prediction performance and significantly improves causal estimation.
This paper explores how imperfect reward models can improve online RLHF.
problem Sample efficiency in online RLHF from imperfect reward models.
method Identifies policy coverability and proposes TPO algorithm for transfer learning.
result TPO algorithm improves sample efficiency compared to standard online learning.
Paper proposes a new trading strategy using corporate event detection from news articles.
problem Predicting stock movements based on corporate events from news articles.
method Bi-level event detection model: low-level for token-level event identification, high-level for article-level event identification.
result The proposed strategy outperforms existing models in stock prediction metrics.
Elo rating outperforms complex models in skill estimation despite model misspecification.
problem Elo rating's reliability in skill estimation is questioned due to model misspecification.
method Interpreted Elo as online gradient descent and conducted synthetic experiments on non-BT models.
result Elo's superior performance in win rate prediction compared to complex models is explained through its sparsity and ranking effectiveness.
Attackers can significantly reduce team rewards in cooperative multi-agent reinforcement learning.
problem Robustness of cooperative multi-agent reinforcement learning to adversaries.
method Novel attack method involving training a policy network and using targeted adversarial examples.
result Reduces team reward from 20 to 9.4 by attacking a single agent, reducing winning rate from 98.9% to 0%.
EBBS integrates expert assessments into MIO best-subsets problem.
problem Identifying relevant features in statistical models.
method Expert-implied Bayesian approach using MIO.
result Reduces to classical best-subsets when experts are silent.
A method to improve LLMs by automating the construction of a mixture of expert prompts.
problem Limitation of single instruction prompts in covering complex problem spaces.
method Divide the problem space into sub-regions, each governed by a specialized expert with both an instruction and demos. A two-phase process constructs these experts.
result Achieves an average win rate of 81% across major benchmarks.
CollaQ improves multi-agent performance in StarCraft by 40% with fewer samples.
problem Limited generalization and high training rounds in multi-agent reinforcement learning.
method Formulates multi-agent collaboration as joint optimization on reward assignment, decomposes Q-function into self and interactive terms, and uses MARA loss.
result Improves win rate by 40% in StarCraft maps with same number of samples compared to state-of-the-art techniques.
Hour-Aware Adaptive Risk Management for Autonomous Memecoin Trading
problem Autonomous memecoin trading system performance
method Hour-of-day effects, filter precision, fragility
result 40.5% win rate, mean per-trade return of +0.62%, cumulative +117.7%
eva method improves RL models by adaptively creating prompts, boosting performance.
problem Fixed prompt distribution limits scalability of RL models post-training.
method Evolving Alignment via Asymmetric Self-Play (eva) approach.
result Significant performance boost on benchmarks, e.g. 51.6% to 60.1% win-rate.
Hybrid method uses LLM to filter lead-lag relationships in prediction markets.
problem Challenges in discovering robust lead-lag relationships in prediction markets due to spurious correlations.
method Two-stage approach: statistical Granger causality followed by LLM semantic re-ranking.
result LLM-based method outperforms statistical baseline, increasing win rate and reducing average loss magnitude.
Recent progress in artificial intelligence through reinforcement learning (RL) has shown great success on increasingly complex single-agent environments and two-player turn-based games. However, the real-world contains multiple agents, each learning and acting independently to cooperate and compete with other agents, a…
This study proposes a framework for identifying profitable trading opportunities based on volatility and causal relationships.
problem Identifying profitable trading opportunities in financial markets.
method A combination of Gaussian Mixture Model (GMM), Granger Causality Test (GCT), Peter-Clark Momentary Conditional Independence (PCMCI) test, Dynamic Time Warping (DTW), and K-Nearest Neighbours (KNN) for identifying and executing trades.
result The proposed volatility-based trading strategy outperformed a Buy-and-Hold strategy, yielding a total return of 15.38%.
SRPO improves AI alignment with human preferences through self-improvement and task-independent optimization.
problem AI models trained with RLHF lack self-correction mechanisms and struggle with task generalization.
method SRPO formulates the preference learning problem as a min-max objective, optimizing a self-improvement policy and a generative policy in an adversarial fashion, making the solution task-independent.
result SRPO outperforms existing methods, achieving 90% AI Win-Rate on XSum and 56% on Arena-Hard prompts after a single revision.