Bidirectional diffusion models predict their own rollout errors without ground truth.
problem Long rollouts accumulate error in autoregressive models, lacking a reliable test-time error signal.
method Train a bidirectional latent diffusion model that steps forward or backward, measuring discrepancies to estimate error.
result Bidirectional consistency Ci ranks rollout error and predicts magnitude with high accuracy. A new method reduces compounding errors in model-based reinforcement learning.
problem Compounding errors in long horizon predictions from model-based reinforcement learning.
method Maximum Entropy Model Rollouts (MEMR) with non-uniform sampling and prioritized experience replay.
result Significantly reduces computation requirements compared to other model-based methods.
This study improves lookahead Bayesian optimization using rollout approximation.
problem Error propagation in lookahead Bayesian optimization due to model mis-specification.
method Proves rollout's improving nature in lookahead BO and provides guidelines for choosing the rolling horizon.
result Empirical results show rollout improves over myopic and non-myopic BO algorithms.
Efficiently reduces computational burden of rollout acquisition functions in Bayesian optimization.
problem Expensive computation of rollout acquisition functions in Bayesian optimization.
method Combines quasi-Monte Carlo, common random numbers, and control variates to reduce computational burden. Formulates a policy-search approach to eliminate the need to optimize the rollout acquisition function.
result Significant reduction in computational burden of rollout acquisition functions.
A new estimator combines bootstrapping and rollout methods in RL.
problem Combining strengths of bootstrapping and rollout methods in RL.
method Subgraph Bellman operators and fixed point solving.
result Upper bound on error approaches optimal TD variance with additional term.
Theory for RLHF generalization under reward shift and clipped KL.
problem Theoretical understanding of RLHF generalization, especially with reward shift and clipped KL.
method Developed generalization theory for RLHF, accounting for reward shift and clipped KL.
result Presented generalization bounds for RLHF, suggesting generalization error from sampling, reward shift, and KL clipping.
Proposes a method to improve model-based reinforcement learning by matching multi-step rollout distributions.
problem Mismatching between model-based and real-world rollouts due to estimation error.
method Uses Wasserstein GAN (WGAN) to match the distributions of multi-step rollouts from the learned model to the real ones.
result The method minimizes the difference in cumulative rewards between the real and learned models.
BASIS improves LLM reasoning by sharing batchwise rollout info, reducing MSE by 69%.
problem Improving large language model reasoning with limited rollouts and batch information.
method BASIS samples only one rollout per prompt but uses batch information to improve value function estimation.
result BASIS reduces MSE in value function estimation by 69% compared to REINFORCE++.
New method finds unseen states for RL, improving performance.
problem Offline RL struggles with unseen states and actions.
method Value-informed state perturbations and filtering.
result Improved performance in offline RL tasks.
Thermalizer stabilizes autoregressive models for long-term predictions in chaotic systems.
problem Long-term predictions in chaotic spatiotemporal systems are unreliable due to trajectory divergence.
method Diffusion models are used to implicitly estimate the score of an invariant measure, which stabilizes autoregressive emulators by applying denoising during inference.
result Thermalization extends the time horizon of stable predictions by an order of magnitude in chaotic systems.
A new method for faster neural network responses reduces training and inference times.
problem Efficiently integrating temporal features in deep neural networks for real-time applications.
method Theoretical framework and demonstration of different rollouts, including streaming rollout, to achieve faster responses and parallel execution.
result The streaming rollout maximizes early and frequent responses, leading to better performance and reduced runtime.
SUPAID automates vehicle rollout decisions for fleet managers.
problem Automating supervisor's vehicle rollout decisions to prevent costly mistakes.
method Rule mining approach based on 'one-way efficiency' assumption.
result SUPAID significantly reduces costs in a real transit agency setting.
Paper proposes a forecasting solution for network-rollout planning.
problem Accurate estimation of milestone completion times for network-rollout planning.
method Partition-based regression models incorporating data-driven statistical models.
result The proposed approach outperforms alternative models in terms of performance and model complexity.
Optimizes staggered treatment rollouts to minimize cost and error.
problem Efficiently scheduling treatment initiation times for staggered rollouts.
method Non-adaptive and adaptive experimental designs, including a near-optimal solution for non-adaptive cases and a new algorithm for adaptive cases.
result Reduces experiment cost by over 50% compared to static design benchmarks.
A contraction analysis improves model-based RL's error recovery.
problem Theoretical understanding of model-based reinforcement learning.
method Contraction analysis applied to both stochastic and deterministic state transitions.
result Error reduction in cumulative reward using branched rollouts.
Adaptive correlated MC improves sequence generation stability.
problem High gradient variance in reinforcement learning for sequence generation.
method Adapts policy gradient estimator using correlated Monte Carlo rollouts.
result Reduces gradient variance and improves model performance.
AIS corrects rollout-training mismatch in quantized RL, improving speed and stability.
problem Rollout-training mismatch in quantized RL causes bias and training collapse.
method Adaptive Importance Sampling (AIS) adjusts gradient correction per batch.
result AIS matches BF16 baseline on most tasks while improving speed.
WALL-E accelerates reinforcement learning by parallelizing experience collection.
problem Slow experience collection time in reinforcement learning.
method Parallel rollout samplers to speed up experience generation.
result Achieved higher average rewards with parallel processes.
Study on indexability of restless multi-armed bandits and rollout policy performance.
problem Maximizing discounted rewards in finite state restless multi-armed bandit problems.
method Decouple the problem into single-armed restless bandits, analyze using value iteration, and compare with Whittle index policy.
result Demonstrates conditions for indexability and compares performance of index policy and rollout policy.
Optimal sample complexity for autoregressive chain-of-thought learning proven.
problem Determining the minimum number of samples needed for accurate autoregressive chain-of-thought learning.
method Proved upper bound on sample complexity using Daniely-Shalev-Shwartz dimension and roll-out stable parity dimension.
result The sample complexity is bounded by the local next-token class rate, with no dependence on rollout length.
SFPO optimizes LLM reasoning by repositioning before updating, improving stability and efficiency.
problem Noisy gradients from low-quality rollouts cause instability and inefficient exploration in on-policy RL algorithms.
method Decomposes each step into three stages: a short fast trajectory, repositioning, and slow correction, preserving the objective and rollout process unchanged.
result SFPO consistently improves stability, reduces rollouts, and accelerates convergence, outperforming GRPO on math reasoning benchmarks.
Generative model learns to mimic time-series behavior to generate accurate trajectories.
problem Learning generative models for time-series data with accurate multi-step trajectories.
method Contrastive imitation framework combining autoregressive and adversarial elements.
result The method generates accurate and useful samples from real-world datasets.
HS-FNO models non-Markovian PDEs by learning history and future states.
problem Non-Markovian dynamics where future states depend on past history.
method History-Space Fourier Neural Operator (HS-FNO) for delay and memory-driven PDEs.
result HS-FNO achieves lowest aggregate errors across various PDE families.
Improves data efficiency in multi-agent control tasks using model-based reinforcement learning.
problem Limited data efficiency in reinforcement learning for multi-agent tasks.
method Decentralized model-based policy optimization (DMPO) framework.
result DMPO achieves superior data efficiency and matches model-free methods using true models.
CoDistill-GRPO improves small models in GRPO by distilling knowledge from a larger model.
problem Small models in GRPO struggle with sparse rewards on difficult tasks.
method Simultaneously trains a large and small model using co-distillation and GRPO objectives.
result Significant improvement in small model performance over standard GRPO on mathematical benchmarks.
Study examines how COVID-19 vaccine companies' popularity affects their stock prices.
problem Impact of COVID-19 vaccine development and rollout on stock prices and company popularity.
method Used Python and various libraries to analyze Google Trends data and stock prices of five vaccine companies.
result Significant correlation between Google Trend data and stock prices, with post-rollout periods showing a slight negative correlation.
SyMetric evaluates learned Hamiltonian dynamics from images, improving model stability and interpretability.
problem Lack of reliable metrics to assess learned Hamiltonian dynamics from images.
method Developed SyMetric, a binary indicator based on Hamiltonian dynamics properties.
result SyMetric identifies architectural improvements for better dynamics learning.
A deep RL method using a bottleneck state improves data efficiency.
problem Lack of data-efficiency in deep reinforcement learning.
method Model-based approach combining learned transition model and rollout simulations.
result The Bottleneck Simulator achieves excellent performance on natural language tasks.
CW-EDMD improves prediction accuracy by learning local Koopman models for different state-space regions.
problem Inefficient global Koopman operator approximation for distinct local dynamics.
method Cluster-Weighted EDMD (CW-EDMD) learns a soft phase-space partition and per-cluster EDMD operators using EM objective.
result CW-EDMD significantly reduces prediction errors across various systems and configurations.
HSR reduces analyst earnings forecast errors by lowering travel friction.
problem How HSR connectivity affects analyst earnings forecast errors in China.
method Firm-year panel data from 2008-2019; placebo test to rule out pre-existing trends.
result HSR reduces analyst earnings forecast errors after connectivity, not before.
InfoTree improves reinforcement learning by optimizing tool use with a greedy submodular approach.
problem Maximizing information from tool use in reinforcement learning with limited resources.
method Formalizes Rollout Informativeness, recasts state selection as submodular maximization, and uses UUCB and ABA.
result InfoTree outperforms existing methods across various benchmarks, improving performance by 18.2% on average.
This paper introduces a simple, general framework for likelihood-free Bayesian reinforcement learning, through Approximate Bayesian Computation (ABC). The main advantage is that we only require a prior distribution on a class of simulators (generative models). This is useful in domains where an analytical probabilistic…
UK's rapid vaccine rollout linked to reduced COVID-19 mortality.
problem Assessing the impact of accelerated vaccine rollout on public health outcomes.
method Flexible probabilistic models combining interrupted time series analysis and synthetic control methods with multi-output Gaussian processes.
result Substantial reduction in COVID-19 mortality with little effect on transmission rates.
sGPO optimizes RLVR training by balancing inference FLOPs and training efficiency.
problem RLVR training allocates rollout budget without considering query difficulty.
method sGPO trades inference FLOPs for reduced training FLOPs.
result sGPO matches or exceeds baseline performance while reducing training compute by a factor of three.
Several approximate policy iteration schemes without value functions, which focus on policy representation using classifiers and address policy learning as a supervised learning problem, have been proposed recently. Finding good policies with such methods requires not only an appropriate classifier, but also reliable e…
STEVE improves sample efficiency in reinforcement learning.
problem Combining model-free and model-based reinforcement learning with low sample complexity.
method Dynamic interpolation between model rollouts of various horizon lengths.
result STEVE achieves an order-of-magnitude increase in sample efficiency.
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.
This paper optimizes sampling policies for Bayesian optimization to improve exploration and exploitation.
problem Improving the balance between exploration and exploitation in Bayesian optimization.
method Developed efficient methods to estimate and optimize non-myopic acquisition functions using rollout policies and stochastic gradient optimization.
result Efficient optimization of sampling policies leads to better performance in Bayesian optimization.
UWM-JEPA predicts future scenarios in belief space, improving accuracy in partially observed environments.
problem Predicting future scenarios in partially observed environments with uncertainty.
method Introduces UWM-JEPA, a JEPA world model with a density-matrix latent and learned unitary predictor.
result UWM-JEPA achieves 0.77 accuracy on a hidden-velocity indicator task, outperforming LSTM-JEPA.
Bayesian inference models failure distributions in autonomous systems.
problem Estimating the distribution of failures in complex systems.
method Bayesian inference using system dynamics rollouts and gradient computation.
result Improves sample efficiency and parameter space coverage in autonomous systems.
A new MAB method improves news headline testing efficiency.
problem Optimizing news headlines for increased engagement.
method Batched Multi-Armed Bandit (bTS) with Thompson Sampling.
result The bTS method outperforms test-rollout by 3.69% in clicks.
The use of Reinforcement Learning in real-world scenarios is strongly limited by issues of scale. Most RL learning algorithms are unable to deal with problems composed of hundreds or sometimes even dozens of possible actions, and therefore cannot be applied to many real-world problems. We consider the RL problem in the…
Paper proposes BMPO to optimize policies using bidirectional models.
problem Model-based reinforcement learning's reliance on forward model accuracy.
method Develops BMPO using both forward and backward models for policy optimization.
result BMPO outperforms state-of-the-art methods in sample efficiency and asymptotic performance.
TRM improves long-horizon LLM RL by masking divergent sequences.
problem Long-horizon reinforcement learning with LLMs suffers from off-policy mismatch and approximation errors.
method Derives and applies trust region bounds to control divergence, proposing Trust Region Masking.
result First non-vacuous monotonic improvement guarantees for long-horizon LLM-RL.
TRM improves long-horizon reinforcement learning for LLMs by masking divergent sequences.
problem Long-horizon reinforcement learning for LLMs suffers from off-policy mismatch and approximation errors.
method Derives and applies trust region bounds to control divergence, proposing Trust Region Masking.
result First non-vacuous monotonic improvement guarantees for long-horizon LLM-RL.
Single autoregressive model outperforms ensemble methods in offline reinforcement learning.
problem Offline reinforcement learning with limited data and model errors.
method Infer system dynamics from data and optimize policies on model rollouts, using a single autoregressive model.
result Single autoregressive model achieves better performance than ensembles on the D4RL benchmark.
TradeFM learns market microstructure from trade events, improving financial model accuracy.
problem Lack of generalizable models for market microstructure.
method Generative Transformer model trained on billions of trade events, using scale-invariant features and universal tokenization.
result TradeFM generates rollouts that match key stylized facts of financial returns and outperforms existing models.
WU-UCT parallelizes MCTS with linear speedup and limited performance loss.
problem Challenges in parallelizing Monte Carlo Tree Search (MCTS) due to its sequential nature.
method Introduces unobserved samples to track incomplete simulations and modify UCT tree policy.
result Achieves linear speedup and only limited performance loss with increasing parallel workers.