Ad hoc teamwork problem describes situations where an agent has to cooperate with previously unseen agents to achieve a common goal. For an agent to be successful in these scenarios, it has to have a suitable cooperative skill. One could implement cooperative skills into an agent by using domain knowledge to design the agent's behavior. However, in complex domains, domain knowledge might not be available. Therefore, it is worthwhile to explore how to directly learn cooperative skills from data. In this work, we apply meta-reinforcement learning (meta-RL) formulation in the context of the ad hoc teamwork problem. Our empirical results show that such a method could produce robust cooperative agents in two cooperative environments with different cooperative circumstances: social compliance and language interpretation. (This is a full paper of the extended abstract version.)

Reinforcement learning can train policies that effectively perform complex tasks. However for long-horizon tasks, the performance of these methods degrades with horizon, often necessitating reasoning over and composing lower-level skills. Hierarchical reinforcement learning aims to enable this by providing a bank of low-level skills as action abstractions. Hierarchies can further improve on this by abstracting the space states as well. We posit that a suitable state abstraction should depend on the capabilities of the available lower-level policies. We propose Value Function Spaces: a simple approach that produces such a representation by using the value functions corresponding to each lower-level skill. These value functions capture the affordances of the scene, thus forming a representation that compactly abstracts task relevant information and robustly ignores distractors. Empirical evaluations for maze-solving and robotic manipulation tasks demonstrate that our approach improves long-horizon performance and enables better zero-shot generalization than alternative model-free and model-based methods.

Rising costs in recent years of developing new drugs and treatments have led to extensive research in optimization techniques in biomolecular design. Currently, the most widely used approach in biomolecular design is directed evolution, which is a greedy hill-climbing algorithm that simulates biological evolution. In this paper, we propose a new benchmark of applying reinforcement learning to RNA sequence design, in which the objective function is defined to be the free energy in the sequence's secondary structure. In addition to experimenting with the vanilla implementations of each reinforcement learning algorithm from standard libraries, we analyze variants of each algorithm in which we modify the algorithm's reward function and tune the model's hyperparameters. We show results of the ablation analysis that we do for these algorithms, as well as graphs indicating the algorithm's performance across batches and its ability to search the possible space of RNA sequences. We find that our DQN algorithm performs by far the best in this setting, contrasting with, in which PPO performs the best among all tested algorithms. Our results should be of interest to those in the biomolecular design community and should serve as a baseline for future experiments involving machine learning in molecule design.

A longstanding goal in reinforcement learning is to build intelligent agents that show fast learning and a flexible transfer of skills akin to humans and animals. This paper investigates the integration of two frameworks for tackling those goals: episodic control and successor features. Episodic control is a cognitively inspired approach relying on episodic memory, an instance-based memory model of an agent's experiences. Meanwhile, successor features and generalized policy improvement (SF&GPI) is a meta and transfer learning framework allowing to learn policies for tasks that can be efficiently reused for later tasks which have a different reward function. Individually, these two techniques have shown impressive results in vastly improving sample efficiency and the elegant reuse of previously learned policies. Thus, we outline a combination of both approaches in a single reinforcement learning framework and empirically illustrate its benefits.

Fog computing is introduced by shifting cloud resources towards the users' proximity to mitigate the limitations possessed by cloud computing. Fog environment made its limited resource available to a large number of users to deploy their serverless applications, composed of several serverless functions. One of the primary intentions behind introducing the fog environment is to fulfil the demand of latency and location-sensitive serverless applications through its limited resources. The recent research mainly focuses on assigning maximum resources to such applications from the fog node and not taking full advantage of the cloud environment. This introduces a negative impact in providing the resources to a maximum number of connected users. To address this issue, in this paper, we investigated the optimum percentage of a user's request that should be fulfilled by fog and cloud. As a result, we proposed DeF-DReL, a Systematic Deployment of Serverless Functions in Fog and Cloud environments using Deep Reinforcement Learning, using several real-life parameters, such as distance and latency of the users from nearby fog node, user's priority, the priority of the serverless applications and their resource demand, etc. The performance of the DeF-DReL algorithm is further compared with recent related algorithms. From the simulation and comparison results, its superiority over other algorithms and its applicability to the real-life scenario can be clearly observed.

Abstract-- Dexterous manipulation of arbitrary objects, a fundamental daily task for humans, has been a grand challenge for autonomous robotic systems. Although data-driven approaches using reinforcement learning can develop specialist policies that discover behaviors to control a single object, they often exhibit poor generalization to unseen ones. In this work, we show that policies learned by existing reinforcement learning algorithms can in fact be generalist when combined with multi-task learning and a well-chosen object representation. We show that a single generalist policy can perform in-hand manipulation of over 100 geometrically-diverse realworld objects and generalize to new objects with unseen shape or size. Interestingly, we find that multi-task learning with object point cloud representations not only generalizes better but even outperforms the single-object specialist policies on both training as well as held-out test objects.

Reinforcement learning (RL) requires access to a reward function that incentivizes the right behavior, but these are notoriously hard to specify for complex tasks. Preference-based RL provides an alternative: learning policies using a teacher's preferences without pre-defined rewards, thus overcoming concerns associated with reward engineering. However, it is difficult to quantify the progress in preference-based RL due to the lack of a commonly adopted benchmark. In this paper, we introduce B-Pref: a benchmark specially designed for preference-based RL. A key challenge with such a benchmark is providing the ability to evaluate candidate algorithms quickly, which makes relying on real human input for evaluation prohibitive. At the same time, simulating human input as giving perfect preferences for the ground truth reward function is unrealistic. B-Pref alleviates this by simulating teachers with a wide array of irrationalities, and proposes metrics not solely for performance but also for robustness to these potential irrationalities. We showcase the utility of B-Pref by using it to analyze algorithmic design choices, such as selecting informative queries, for state-of-the-art preference-based RL algorithms. We hope that B-Pref can serve as a common starting point to study preference-based RL more systematically. Source code is available at https://github.com/rll-research/B-Pref.

The rapid advancements of Internet of Things (IoT) and artificial intelligence (AI) have catalyzed the development of adaptive traffic signal control systems (ATCS) for smart cities. In particular, deep reinforcement learning (DRL) methods produce the state-of-the-art performance and have great potentials for practical applications. In the existing DRL-based ATCS, the controlled signals collect traffic state information from nearby vehicles, and then optimal actions (e.g., switching phases) can be determined based on the collected information. The DRL models fully "trust" that vehicles are sending the true information to the signals, making the ATCS vulnerable to adversarial attacks with falsified information. In view of this, this paper first time formulates a novel task in which a group of vehicles can cooperatively send falsified information to "cheat" DRL-based ATCS in order to save their total travel time. To solve the proposed task, we develop CollusionVeh, a generic and effective vehicle-colluding framework composed of a road situation encoder, a vehicle interpreter, and a communication mechanism. We employ our method to attack established DRL-based ATCS and demonstrate that the total travel time for the colluding vehicles can be significantly reduced with a reasonable number of learning episodes, and the colluding effect will decrease if the number of colluding vehicles increases. Additionally, insights and suggestions for the real-world deployment of DRL-based ATCS are provided. The research outcomes could help improve the reliability and robustness of the ATCS and better protect the smart mobility systems.

This is technically Deep Learning in Python part 11 of my deep learning series, and my 3rd reinforcement learning course. Deep Reinforcement Learning is actually the combination of 2 topics: Reinforcement Learning and Deep Learning (Neural Networks). While both of these have been around for quite some time, it's only been recently that Deep Learning has really taken off, and along with it, Reinforcement Learning. The maturation of deep learning has propelled advances in reinforcement learning, which has been around since the 1980s, although some aspects of it, such as the Bellman equation, have been for much longer. Recently, these advances have allowed us to showcase just how powerful reinforcement learning can be.

We propose a general and systematic strategy to compile arbitrary quantum channels without using ancillary qubits, based on proximal policy optimization -- a powerful deep reinforcement learning algorithm. We rigorously prove that, in sharp contrast to the case of compiling unitary gates, it is impossible to compile an arbitrary channel to arbitrary precision with any given finite elementary channel set, regardless of the length of the decomposition sequence. However, for a fixed accuracy $\epsilon$ one can construct a universal set with constant number of $\epsilon$-dependent elementary channels, such that an arbitrary quantum channel can be decomposed into a sequence of these elementary channels followed by a unitary gate, with the sequence length bounded by $O(\frac{1}{\epsilon}\log\frac{1}{\epsilon})$. Through a concrete example concerning topological compiling of Majorana fermions, we show that our proposed algorithm can conveniently and effectively reduce the use of expensive elementary gates through adding the weighted cost into the reward function of the proximal policy optimization.