As representation learning becomes a powerful technique to reduce sample complexity in reinforcement learning (RL) in practice, theoretical understanding of its advantage is still limited. In this paper, we theoretically characterize the benefit of representation learning under the low-rank Markov decision process (MDP) model. We first study multitask low-rank RL (as upstream training), where all tasks share a common representation, and propose a new multitask reward-free algorithm called REFUEL. REFUEL learns both the transition kernel and the near-optimal policy for each task, and outputs a well-learned representation for downstream tasks. Our result demonstrates that multitask representation learning is provably more sample-efficient than learning each task individually, as long as the total number of tasks is above a certain threshold. We then study the downstream RL in both online and offline settings, where the agent is assigned with a new task sharing the same representation as the upstream tasks. For both online and offline settings, we develop a sample-efficient algorithm, and show that it finds a near-optimal policy with the suboptimality gap bounded by the sum of the estimation error of the learned representation in upstream and a vanishing term as the number of downstream samples becomes large. Our downstream results of online and offline RL further capture the benefit of employing the learned representation from upstream as opposed to learning the representation of the low-rank model directly. To the best of our knowledge, this is the first theoretical study that characterizes the benefit of representation learning in exploration-based reward-free multitask RL for both upstream and downstream tasks.

We study reinforcement learning for partially observed Markov decision processes (POMDPs) with infinite observation and state spaces, which remains less investigated theoretically. To this end, we make the first attempt at bridging partial observability and function approximation for a class of POMDPs with a linear structure. In detail, we propose a reinforcement learning algorithm (Optimistic Exploration via Adversarial Integral Equation or OP-TENET) that attains an $\epsilon$-optimal policy within $O(1/\epsilon^2)$ episodes. In particular, the sample complexity scales polynomially in the intrinsic dimension of the linear structure and is independent of the size of the observation and state spaces. The sample efficiency of OP-TENET is enabled by a sequence of ingredients: (i) a Bellman operator with finite memory, which represents the value function in a recursive manner, (ii) the identification and estimation of such an operator via an adversarial integral equation, which features a smoothed discriminator tailored to the linear structure, and (iii) the exploration of the observation and state spaces via optimism, which is based on quantifying the uncertainty in the adversarial integral equation.

Reinforcement Learning is a subfield of Machine Learning, but is also a general purpose formalism for automated decision-making and AI. This course introduces you to statistical learning techniques where an agent explicitly takes actions and interacts with the world. Understanding the importance and challenges of learning agents that make decisions is of vital importance today, with more and more companies interested in interactive agents and intelligent decision-making. This course introduces you to the fundamentals of Reinforcement Learning. When you finish this course, you will: - Formalize problems as Markov Decision Processes - Understand basic exploration methods and the exploration/exploitation tradeoff - Understand value functions, as a general-purpose tool for optimal decision-making - Know how to implement dynamic programming as an efficient solution approach to an industrial control problem This course teaches you the key concepts of Reinforcement Learning, underlying classic and modern algorithms in RL.

DDPG (Deep Deterministic Policy Gradient) is a model-free off-policy reinforcement learning algorithm for learning continuous actions. It combines ideas from DPG (Deterministic Policy Gradient) and DQN (Deep Q-Network). The development of deep deterministic policy gradient (DDPG) was inspired by the success of DQN and is aimed to improve performance for tasks that requires a continuous action space. DDPG incorporates an actor-critic approach based on DPG. The algorithm uses two neural networks, one for the actor and one for the critic.

We derive a concentration bound for a Q-learning algorithm for average cost Markov decision processes based on an equivalent shortest path problem, and compare it numerically with the alternative scheme based on relative value iteration.

In this tutorial, we will be discussing what is Q learning and how to Implement Q learning using Tensorflow2. Q Learning is one of the most popular reinforcement learning algorithms, it is an off-policy reinforcement learning(RL) that finds the best action for the given state. Q learning is considered off-policy reinforcement learning because Q learning is not dependent on current policy. It learns from actions that are outside the current policy, like taking random actions, therefore the policy is not needed. Here Q stands for Quality, which means how useful the given action is for the current state to get some future reward.

We are looking to expand our Analytics team with a data scientist to help create data-driven internal solutions for the Illuvium DAO. Illuvium Labs is an independent game development studio based in Sydney, Australia. We have developed a strong culture of independence with our team, preferring candidates who can articulate their own vision and goals. We operate almost entirely remotely so each team member designs their own hours and work schedule. In the end all that matters is the delivered product.

Offline or batch reinforcement learning seeks to learn a near-optimal policy using history data without active exploration of the environment. To counter the insufficient coverage and sample scarcity of many offline datasets, the principle of pessimism has been recently introduced to mitigate high bias of the estimated values. While pessimistic variants of model-based algorithms (e.g., value iteration with lower confidence bounds) have been theoretically investigated, their model-free counterparts -- which do not require explicit model estimation -- have not been adequately studied, especially in terms of sample efficiency. To address this inadequacy, we study a pessimistic variant of Q-learning in the context of finite-horizon Markov decision processes, and characterize its sample complexity under the single-policy concentrability assumption which does not require the full coverage of the state-action space. In addition, a variance-reduced pessimistic Q-learning algorithm is proposed to achieve near-optimal sample complexity. Altogether, this work highlights the efficiency of model-free algorithms in offline RL when used in conjunction with pessimism and variance reduction.

Proportional-integral-derivative (PID) controller is widely used across various industrial process control applications because of its straightforward implementation. However, it can be challenging to fine-tune the PID parameters in practice to achieve robust performance. The paper proposes a model-based reinforcement learning (RL) framework to design PID controllers leveraging the probabilistic inference for learning control (PILCO) method and Kullback-Leibler divergence (KLD). Since PID controllers have a much more interpretable control structure than a network basis function, an optimal policy given by PILCO is transformed into a set of robust PID tuning parameters for underactuated mechanical systems. The presented method is general and can blend with several model-based and model-free algorithms. The performance of the devised PID controllers is demonstrated with simulation studies for a benchmark cart-pole system under disturbances and system parameter uncertainties.

Information-directed sampling (IDS) has recently demonstrated its potential as a data-efficient reinforcement learning algorithm. However, it is still unclear what is the right form of information ratio to optimize when contextual information is available. We investigate the IDS design through two contextual bandit problems: contextual bandits with graph feedback and sparse linear contextual bandits. We provably demonstrate the advantage of contextual IDS over conditional IDS and emphasize the importance of considering the context distribution. The main message is that an intelligent agent should invest more on the actions that are beneficial for the future unseen contexts while the conditional IDS can be myopic. We further propose a computationally-efficient version of contextual IDS based on Actor-Critic and evaluate it empirically on a neural network contextual bandit.