We propose DRAGO, a novel approach for continual model-based reinforcement learning aimed at improving the incremental development of world models across a sequence of tasks that differ in their reward functions but not the state space or dynamics. DRAGO comprises two key components: Synthetic Experience Rehearsal, which leverages generative models to create synthetic experiences from past tasks, allowing the agent to reinforce previously learned dynamics without storing data, and Regaining Memories Through Exploration, which introduces an intrinsic reward mechanism to guide the agent toward revisiting relevant states from prior tasks. Together, these components enable the agent to maintain a comprehensive and continually developing world model, facilitating more effective learning and adaptation across diverse environments. Empirical evaluations demonstrate that DRAGO is able to preserve knowledge across tasks, achieving superior performance in various continual learning scenarios.

The complexity of online decision-making under uncertainty stems from the requirement of finding a balance between exploiting known strategies and exploring new possibilities. Naturally, the uncertainty type plays a crucial role in developing decision-making strategies that manage complexity effectively. In this paper, we focus on a specific form of uncertainty known as epistemic ambivalence (EA), which emerges from conflicting pieces of evidence or contradictory experiences. It creates a delicate interplay between uncertainty and confidence, distinguishing it from epistemic uncertainty that typically diminishes with new information. Indeed, ambivalence can persist even after additional knowledge is acquired. To address this phenomenon, we propose a novel framework, called the epistemically ambivalent Markov decision process (EA-MDP), aiming to understand and control EA in decision-making processes. This framework incorporates the concept of a quantum state from the quantum mechanics formalism, and its core is to assess the probability and reward of every possible outcome. We calculate the reward function using quantum measurement techniques and prove the existence of an optimal policy and an optimal value function in the EA-MDP framework. We also propose the EA-epsilon-greedy Q-learning algorithm. To evaluate the impact of EA on decision-making and the expedience of our framework, we study two distinct experimental setups, namely the two-state problem and the lattice problem. Our results show that using our methods, the agent converges to the optimal policy in the presence of EA.

Portfolio management remains a crucial challenge in finance, with traditional methods often falling short in complex and volatile market environments. While deep reinforcement approaches have shown promise, they still face limitations in dynamic risk management, exploitation of temporal markets, and incorporation of complex trading strategies such as short-selling. These limitations can lead to suboptimal portfolio performance, increased vulnerability to market volatility, and missed opportunities in capturing potential returns from diverse market conditions. This paper introduces a Deep Reinforcement Learning Portfolio Management Framework with Time-Awareness and Short-Selling (MTS), offering a robust and adaptive strategy for sustainable investment performance. This framework utilizes a novel encoder-attention mechanism to address the limitations by incorporating temporal market characteristics, a parallel strategy for automated short-selling based on market trends, and risk management through innovative Incremental Conditional Value at Risk, enhancing adaptability and performance. Experimental validation on five diverse datasets from 2019 to 2023 demonstrates MTS's superiority over traditional algorithms and advanced machine learning techniques. MTS consistently achieves higher cumulative returns, Sharpe, Omega, and Sortino ratios, underscoring its effectiveness in balancing risk and return while adapting to market dynamics. MTS demonstrates an average relative increase of 30.67% in cumulative returns and 29.33% in Sharpe ratio compared to the next best-performing strategies across various datasets.

Designing effective reward functions in multi-agent reinforcement learning (MARL) is a significant challenge, often leading to suboptimal or misaligned behaviors in complex, coordinated environments. We introduce Multi-agent Reinforcement Learning from Multi-phase Human Feedback of Mixed Quality (M3HF), a novel framework that integrates multi-phase human feedback of mixed quality into the MARL training process. By involving humans with diverse expertise levels to provide iterative guidance, M3HF leverages both expert and non-expert feedback to continuously refine agents' policies. During training, we strategically pause agent learning for human evaluation, parse feedback using large language models to assign it appropriately and update reward functions through predefined templates and adaptive weight by using weight decay and performance-based adjustments. Our approach enables the integration of nuanced human insights across various levels of quality, enhancing the interpretability and robustness of multi-agent cooperation. Empirical results in challenging environments demonstrate that M3HF significantly outperforms state-of-the-art methods, effectively addressing the complexities of reward design in MARL and enabling broader human participation in the training process.

While internet-scale image and textual data have enabled strong generalization in Vision-Language Models (VLMs), the absence of internet-scale control data has impeded the development of similar generalization in standard reinforcement learning (RL) agents. Although VLMs are fundamentally limited in their ability to solve control tasks due to their lack of action-conditioned training data, their capacity for image understanding allows them to provide valuable feedback in RL tasks by recognizing successful outcomes. A key challenge in Reinforcement Learning from AI Feedback (RLAIF) is determining how best to integrate VLM-derived signals into the learning process. We explore this question in the context of offline RL and introduce a class of methods called sub-trajectory filtered optimization. We identify three key insights. First, trajectory length plays a crucial role in offline RL, as full-trajectory preference learning exacerbates the stitching problem, necessitating the use of sub-trajectories. Second, even in Markovian environments, a non-Markovian reward signal from a sequence of images is required to assess trajectory improvement, as VLMs do not interpret control actions and must rely on visual cues over time. Third, a simple yet effective approach--filtered and weighted behavior cloning--consistently outperforms more complex reinforcement learning from human feedback-based methods. We propose sub-trajectory filtered behavior cloning, a method that leverages VLM feedback on sub-trajectories while incorporating a retrospective filtering mechanism that removes sub-trajectories preceding failures to improve robustness and prevent turbulence. This study is preliminary; we provide initial evidence through evaluations on a toy control domain. Please enjoy our airport puns.

Boosting provides a practical and provably effective framework for constructing accurate learning algorithms from inaccurate rules of thumb. It extends the promise of sample-efficient learning to settings where direct Empirical Risk Minimization (ERM) may not be implementable efficiently. In the realizable setting, boosting is known to offer this computational reprieve without compromising on sample efficiency. However, in the agnostic case, existing boosting algorithms fall short of achieving the optimal sample complexity. This paper highlights an unexpected and previously unexplored avenue of improvement: unlabeled samples. We design a computationally efficient agnostic boosting algorithm that matches the sample complexity of ERM, given polynomially many additional unlabeled samples. In fact, we show that the total number of samples needed, unlabeled and labeled inclusive, is never more than that for the best known agnostic boosting algorithm -- so this result is never worse -- while only a vanishing fraction of these need to be labeled for the algorithm to succeed. This is particularly fortuitous for learning-theoretic applications of agnostic boosting, which often take place in the distribution-specific setting, where unlabeled samples can be availed for free. We detail other applications of this result in reinforcement learning.

In the 1980s, Andrew Barto and Rich Sutton were considered eccentric devotees to an elegant but ultimately doomed idea--having machines learn, as humans and animals do, from experience. Decades on, with the technique they pioneered now increasingly critical to modern artificial intelligence and programs like ChatGPT, Barto and Sutton have been awarded the Turing Award, the highest honor in the field of computer science. Barto, a professor emeritus at the University of Massachusetts Amherst, and Sutton, a professor at the University of Alberta, trailblazed a technique known as reinforcement learning, which involves coaxing a computer to perform tasks through experimentation combined with either positive or negative feedback. "When this work started for me, it was extremely unfashionable," Barto recalls with a smile, speaking over Zoom from his home in Massachusetts. "It's been remarkable that [it has] achieved some influence and some attention," Barto adds.

Efficient exploration is essential for intelligent systems interacting with their environment, but existing language models often fall short in scenarios that require strategic information gathering. In this paper, we present PAPRIKA, a fine-tuning approach that enables language models to develop general decision-making capabilities that are not confined to particular environments. By training on synthetic interaction data from different tasks that require diverse strategies, PAPRIKA teaches models to explore and adapt their behavior on a new task based on environment feedback in-context without more gradient updates. Experimental results show that models fine-tuned with PAPRIKA can effectively transfer their learned decision-making capabilities to entirely unseen tasks without additional training. Unlike traditional training, our approach's primary bottleneck lies in sampling useful interaction data instead of model updates. To improve sample efficiency, we propose a curriculum learning strategy that prioritizes sampling trajectories from tasks with high learning potential. These results suggest a promising path towards AI systems that can autonomously solve novel sequential decision-making problems that require interactions with the external world.

We offer a new in-depth investigation of global path planning (GPP) for unmanned ground vehicles, an autonomous mining sampling robot named ROMIE. GPP is essential for ROMIE's optimal performance, which is translated into solving the traveling salesman problem, a complex graph theory challenge that is crucial for determining the most effective route to cover all sampling locations in a mining field. This problem is central to enhancing ROMIE's operational efficiency and competitiveness against human labor by optimizing cost and time. The primary aim of this research is to advance GPP by developing, evaluating, and improving a cost-efficient software and web application. We delve into an extensive comparison and analysis of Google operations research (OR)-Tools optimization algorithms. Our study is driven by the goal of applying and testing the limits of OR-Tools capabilities by integrating Reinforcement Learning techniques for the first time. This enables us to compare these methods with OR-Tools, assessing their computational effectiveness and real-world application efficiency. Our analysis seeks to provide insights into the effectiveness and practical application of each technique. Our findings indicate that Q-Learning stands out as the optimal strategy, demonstrating superior efficiency by deviating only 1.2% on average from the optimal solutions across our datasets.

We investigate efficient exploration strategies of environments with unknown stochastic dynamics and sparse rewards. Specifically, we analyze first the impact of parallel simulations on the probability of reaching rare states within a finite time budget. Using simplified models based on random walks and L\'evy processes, we provide analytical results that demonstrate a phase transition in reaching probabilities as a function of the number of parallel simulations. We identify an optimal number of parallel simulations that balances exploration diversity and time allocation. Additionally, we analyze a restarting mechanism that exponentially enhances the probability of success by redirecting efforts toward more promising regions of the state space. Our findings contribute to a more qualitative and quantitative theory of some exploration schemes in reinforcement learning, offering insights into developing more efficient strategies for environments characterized by rare events.