An official website of the United States government
Reinforcement Learning (RL) is a sampling-based method for sequential decision-making, in which a learning agent iteratively converges toward an optimal policy by leveraging feedback from the environment in the form of scalar reward signals.While timing information is often abstracted in discrete-time domains, time-critical learning applications—such as queuing systems, population processes, and manufacturing systems—are naturally modeled as Continuous-Time Markov Decision Processes (CTMDPs).Since the seminal work of Bradtke and Duff, model-free RL for CTMDPs has become well-understood. However, in many practical applications, practitioners possess high-quality information about system rates derived from traditional queuing theory, which learning agents could potentially exploit to accelerate convergence. Despite this, classical RL algorithms for CTMDPs typically re-learn these parameters through sampling.In this work, we propose continuous-time reward machines (CTRMs), a novel framework that embeds reward functions and real-time state-action dynamics into a unified structure. CTRMs enable RL agents to effectively navigate dense-time environments while leveraging reward shaping and counterfactual experiences for accelerated learning. Our empirical results demonstrate CTRMs' ability to improve learning efficiency in time-critical environments.
more »
« less
Cliff Diving: Exploring Reward Surfaces in Reinforcement Learning Environments
Visualizing optimization landscapes has resulted in many fundamental insights in numeric optimization, specifically regarding novel improvements to optimization techniques. However, visualizations of the objective that reinforcement learning optimizes (the "reward surface") have only ever been generated for a small number of narrow contexts. This work presents reward surfaces and related visualizations of 27 of the most widely used reinforcement learning environments in Gym for the first time. We also explore reward surfaces in the policy gradient direction and show for the first time that many popular reinforcement learning environments have frequent "cliffs" (sudden large drops in expected reward). We demonstrate that A2C often "dives off" these cliffs into low reward regions of the parameter space while PPO avoids them, confirming a popular intuition for PPO’s improved performance over previous methods. We additionally introduce a highly extensible library that allows researchers to easily generate these visualizations in the future. Our findings provide new intuition to explain the successes and failures of modern RL methods, and our visualizations concretely characterize several failure modes of reinforcement learning agents in novel ways.
more »
« less
- PAR ID:
- 10409118
- Date Published:
- Journal Name:
- Proceedings of the 39th International Conference on Machine Learning
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Cache management is a critical aspect of computer architecture, encompassing techniques such as cache replacement, bypassing, and prefetching. Existing research has often focused on individual techniques, overlooking the potential benefits of joint optimization. Moreover, many of these approaches rely on static and intuition-driven policies, limiting their performance under complex and dynamic workloads. To address these challenges, this paper introduces CHROME, a novel concurrencyaware cache management framework. CHROME takes a holistic approach by seamlessly integrating intelligent cache replacement and bypassing with pattern-based prefetching. By leveraging online reinforcement learning, CHROME dynamically adapts cache decisions based on multiple program features and applies a reward for each decision that considers the accuracy of the action and the system-level feedback information. Our performance evaluation demonstrates that CHROME outperforms current state-of-the-art schemes, exhibiting significant improvements in cache management. Notably, CHROME achieves a remarkable performance boost of up to 13.7% over the traditional LRU method in multi-core systems with only modest overhead.more » « less
-
null (Ed.)Learning in tightly coupled multiagent settings with sparse rewards is challenging because multiple agents must reach the goal state simultaneously for the team to receive a reward. This is even more challenging under temporal coupling constraints - where agents need to sequentially complete different components of a task in a particular order. Here, a single local reward is inadequate for learning an effective policy. We introduce MADyS, Multiagent Learning via Dynamic Skill Selection, a bi-level optimization framework that learns to dynamically switch between multiple local skills to optimize sparse team objectives. MADyS adopts fast policy gradients to learn local skills using local rewards and an evolutionary algorithm to optimize the sparse team objective by recruiting the most optimal skill at any given time. This eliminates the need to generate a single dense reward via reward shaping or other mixing functions. In environments with both spatial and temporal coupling requirements, we outperform prior methods and provides intuitive visualizations of its skill switching strategy.more » « less
-
Designing a good reward function is essential to robot planning and reinforcement learning, but it can also be challenging and frustrating. The reward needs to work across multiple different environments, and that often requires many iterations of tuning. We introduce a novel divide-and- conquer approach that enables the designer to specify a reward separately for each environment. By treating these separate reward functions as observations about the underlying true reward, we derive an approach to infer a common reward across all environments. We conduct user studies in an abstract grid world domain and in a motion planning domain for a 7-DOF manipulator that measure user effort and solution quality. We show that our method is faster, easier to use, and produces a higher quality solution than the typical method of designing a reward jointly across all environments. We additionally conduct a series of experiments that measure the sensitivity of these results to different properties of the reward design task, such as the number of environments, the number of feasible solutions per environment, and the fraction of the total features that vary within each environment. We find that independent reward design outperforms the standard, joint, reward design process but works best when the design problem can be divided into simpler subproblems.more » « less
-
Recent reinforcement learning (RL) approaches have shown strong performance in complex domains such as Atari games, but are often highly sample inefficient. A common approach to reduce interaction time with the environment is to use reward shaping, which involves carefully designing reward functions that provide the agent intermediate rewards for progress towards the goal. However, designing appropriate shaping rewards is known to be difficult as well as time-consuming. In this work, we address this problem by using natural language instructions to perform reward shaping. We propose the LanguagE-Action Reward Network (LEARN), a framework that maps free-form natural language instructions to intermediate rewards based on actions taken by the agent. These intermediate language-based rewards can seamlessly be integrated into any standard reinforcement learning algorithm. We experiment with Montezuma’s Revenge from the Atari Learning Environment, a popular benchmark in RL. Our experiments on a diverse set of 15 tasks demonstrate that, for the same number of interactions with the environment, language-based rewards lead to successful completion of the task 60 % more often on average, compared to learning without language.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Conference Paper:
- The DOI is not currently available.
