Memory Gym: Partially Observable Challenges to Memory-Based Agents in Endless Episodes

Memory Gym introduces a unique benchmark designed to test Deep Reinforcement Learning agents, specifically comparing Gated Recurrent Unit (GRU) against Transformer-XL (TrXL), on their ability to memorize long sequences, withstand noise, and generalize. It features partially observable 2D environments with discrete controls, namely Mortar Mayhem, Mystery Path, and Searing Spotlights. These originally finite environments are extrapolated to novel endless tasks that act as an automatic curriculum, drawing inspiration from the car game ``I packed my bag". These endless tasks are not only beneficial for evaluating efficiency but also intriguingly valuable for assessing the effectiveness of approaches in memory-based agents. Given the scarcity of publicly available memory baselines, we contribute an implementation driven by TrXL and Proximal Policy Optimization. This implementation leverages TrXL as episodic memory using a sliding window approach. In our experiments on the finite environments, TrXL demonstrates superior sample efficiency in Mystery Path and outperforms in Mortar Mayhem. However, GRU is more efficient on Searing Spotlights. Most notably, in all endless tasks, GRU makes a remarkable resurgence, consistently outperforming TrXL by significant margins.

Believable Minecraft Settlements by Means of Decentralised Iterative Planning

Procedural city generation that focuses on believability and adaptability to random terrain is a difficult challenge in the field of Procedural Content Generation (PCG). Dozens of researchers compete for a realistic approach in challenges such as the Generative Settlement Design in Minecraft (GDMC), in which our method has won the 2022 competition. This was achieved through a decentralised, iterative planning process that is transferable to similar generation processes that aims to produce "organic" content procedurally.

Models Matter: The Impact of Single-Step Retrosynthesis on Synthesis Planning

Retrosynthesis consists of breaking down a chemical compound recursively step-by-step into molecular precursors until a set of commercially available molecules is found with the goal to provide a synthesis route. Its two primary research directions, single-step retrosynthesis prediction, which models the chemical reaction logic, and multi-step synthesis planning, which tries to find the correct sequence of reactions, are inherently intertwined. Still, this connection is not reflected in contemporary research. In this work, we combine these two major research directions by applying multiple single-step retrosynthesis models within multi-step synthesis planning and analyzing their impact using public and proprietary reaction data. We find a disconnection between high single-step performance and potential route-finding success, suggesting that single-step models must be evaluated within synthesis planning in the future. Furthermore, we show that the commonly used single-step retrosynthesis benchmark dataset USPTO-50k is insufficient as this evaluation task does not represent model performance and scalability on larger and more diverse datasets. For multi-step synthesis planning, we show that the choice of the single-step model can improve the overall success rate of synthesis planning by up to +28% compared to the commonly used baseline model. Finally, we show that each single-step model finds unique synthesis routes, and differs in aspects such as route-finding success, the number of found synthesis routes, and chemical validity, making the combination of single-step retrosynthesis prediction and multi-step synthesis planning a crucial aspect when developing future methods.

Two-Memory Reinforcement Learning

While deep reinforcement learning has shown important empirical success, it tends to learn relatively slow due to slow propagation of rewards information and slow update of parametric neural networks. Non-parametric episodic memory, on the other hand, provides a faster learning alternative that does not require representation learning and uses maximum episodic return as state-action values for action selection. Episodic memory and reinforcement learning both have their own strengths and weaknesses. Notably, humans can leverage multiple memory systems concurrently during learning and benefit from all of them. In this work, we propose a method called Two-Memory reinforcement learning agent (2M) that combines episodic memory and reinforcement learning to distill both of their strengths. The 2M agent exploits the speed of the episodic memory part and the optimality and the generalization capacity of the reinforcement learning part to complement each other. Our experiments demonstrate that the 2M agent is more data efficient and outperforms both pure episodic memory and pure reinforcement learning, as well as a state-of-the-art memory-augmented RL agent. Moreover, the proposed approach provides a general framework that can be used to combine any episodic memory agent with other off-policy reinforcement learning algorithms.

Mind the Retrosynthesis Gap: Bridging the divide between Single-step and Multi-step Retrosynthesis Prediction

Retrosynthesis is the task of breaking down a chemical compound recursively step-by-step into molecular precursors until a set of commercially available molecules is found. Consequently, the goal is to provide a valid synthesis route for a molecule. As more single-step models develop, we see increasing accuracy in the prediction of molecular disconnections, potentially improving the creation of synthetic paths. Multi-step approaches repeatedly apply the chemical information stored in single-step retrosynthesis models. However, this connection is not reflected in contemporary research, fixing either the single-step model or the multi-step algorithm in the process. In this work, we establish a bridge between both tasks by benchmarking the performance and transfer of different single-step retrosynthesis models to the multi-step domain by leveraging two common search algorithms, Monte Carlo Tree Search and Retro*. We show that models designed for single-step retrosynthesis, when extended to multi-step, can have a tremendous impact on the route finding capabilities of current multi-step methods, improving performance by up to +30% compared to the most widely used model. Furthermore, we observe no clear link between contemporary single-step and multi-step evaluation metrics, showing that single-step models need to be developed and tested for the multi-step domain and not as an isolated task to find synthesis routes for molecules of interest.

First Go, then Post-Explore: the Benefits of Post-Exploration in Intrinsic Motivation

Go-Explore achieved breakthrough performance on challenging reinforcement learning (RL) tasks with sparse rewards. The key insight of Go-Explore was that successful exploration requires an agent to first return to an interesting state ('Go'), and only then explore into unknown terrain ('Explore'). We refer to such exploration after a goal is reached as 'post-exploration'. In this paper, we present a clear ablation study of post-exploration in a general intrinsically motivated goal exploration process (IMGEP) framework, that the Go-Explore paper did not show. We study the isolated potential of post-exploration, by turning it on and off within the same algorithm under both tabular and deep RL settings on both discrete navigation and continuous control tasks. Experiments on a range of MiniGrid and Mujoco environments show that post-exploration indeed helps IMGEP agents reach more diverse states and boosts their performance. In short, our work suggests that RL researchers should consider to use post-exploration in IMGEP when possible since it is effective, method-agnostic and easy to implement.

Continuous Episodic Control

Non-parametric episodic memory can be used to quickly latch onto high-reward experience in reinforcement learning tasks. In contrast to parametric deep reinforcement learning approaches, these methods only need to discover the solution once, and may then repeatedly solve the task. However, episodic control solutions are stored in discrete tables, and this approach has so far only been applied to discrete action space problems. Therefore, this paper introduces Continuous Episodic Control (CEC), a novel non-parametric episodic memory algorithm for sequential decision making in problems with a continuous action space. Results on several sparse-reward continuous control environments show that our proposed method learns faster than state-of-the-art model-free RL and memory-augmented RL algorithms, while maintaining good long-run performance as well. In short, CEC can be a fast approach for learning in continuous control tasks, and a useful addition to parametric RL methods in a hybrid approach as well.

On the Verge of Solving Rocket League using Deep Reinforcement Learning and Sim-to-sim Transfer

Autonomously trained agents that are supposed to play video games reasonably well rely either on fast simulation speeds or heavy parallelization across thousands of machines running concurrently. This work explores a third way that is established in robotics, namely sim-to-real transfer, or if the game is considered a simulation itself, sim-to-sim transfer. In the case of Rocket League, we demonstrate that single behaviors of goalies and strikers can be successfully learned using Deep Reinforcement Learning in the simulation environment and transferred back to the original game. Although the implemented training simulation is to some extent inaccurate, the goalkeeping agent saves nearly 100% of its faced shots once transferred, while the striking agent scores in about 75% of cases. Therefore, the trained agent is robust enough and able to generalize to the target domain of Rocket League.

Generalization, Mayhems and Limits in Recurrent Proximal Policy Optimization

At first sight it may seem straightforward to use recurrent layers in Deep Reinforcement Learning algorithms to enable agents to make use of memory in the setting of partially observable environments. Starting from widely used Proximal Policy Optimization (PPO), we highlight vital details that one must get right when adding recurrence to achieve a correct and efficient implementation, namely: properly shaping the neural net's forward pass, arranging the training data, correspondingly selecting hidden states for sequence beginnings and masking paddings for loss computation. We further explore the limitations of recurrent PPO by benchmarking the contributed novel environments Mortar Mayhem and Searing Spotlights that challenge the agent's memory beyond solely capacity and distraction tasks. Remarkably, we can demonstrate a transition to strong generalization in Mortar Mayhem when scaling the number of training seeds, while the agent does not succeed on Searing Spotlights, which seems to be a tough challenge for memory-based agents.