The solution convergence of Markov Decision Processes (MDPs) can be accelerated by prioritized sweeping of states ranked by their potential impacts to other states. In this paper, we present new heuristics to speed up the solution convergence of MDPs. First, we quantify the level of reachability of every state using the Mean First Passage Time (MFPT) and show that such reachability characterization very well assesses the importance of states which is used for effective state prioritization. Then, we introduce the notion of backup differentials as an extension to the prioritized sweeping mechanism, in order to evaluate the impacts of states at an even finer scale. Finally, we extend the state prioritization to the temporal process, where only partial sweeping can be performed during certain intermediate value iteration stages. To validate our design, we have performed numerical evaluations by comparing the proposed new heuristics with corresponding classic baseline mechanisms. The evaluation results showed that our reachability based framework and its differential variants have outperformed the state-of-the-art solutions in terms of both practical runtime and number of iterations.

We investigate whether quantum annealers with select chip layouts can outperform classical computers in reinforcement learning tasks. We associate a transverse field Ising spin Hamiltonian with a layout of qubits similar to that of a deep Boltzmann machine (DBM) and use simulated quantum annealing (SQA) to numerically simulate quantum sampling from this system. We design a reinforcement learning algorithm in which the set of visible nodes representing the states and actions of an optimal policy are the first and last layers of the deep network. In absence of a transverse field, our simulations show that DBMs are trained more effectively than restricted Boltzmann machines (RBM) with the same number of nodes. We then develop a framework for training the network as a quantum Boltzmann machine (QBM) in the presence of a significant transverse field for reinforcement learning. This method also outperforms the reinforcement learning method that uses RBMs.

In this paper, we propose a new deep neural network architecture, called NMD net, that has been specifically designed to learn adaptive behaviours. This architecture exploits a biological mechanism called neuromodulation that sustains adaptation in biological organisms. This architecture has been introduced in a deep-reinforcement learning architecture for interacting with Markov decision processes in a meta-reinforcement learning setting where the action space is continuous. The deep-reinforcement learning architecture is trained using an advantage actor-critic algorithm. Experiments are carried on several test problems. Results show that the neural network architecture with neuromodulation provides significantly better results than state-of-the-art recurrent neural networks which do not exploit this mechanism.

Multi-output learning aims to simultaneously predict multiple outputs given an input. It is an important learning problem due to the pressing need for sophisticated decision making in real-world applications. Inspired by big data, the 4Vs characteristics of multi-output imposes a set of challenges to multi-output learning, in terms of the volume, velocity, variety and veracity of the outputs. Increasing number of works in the literature have been devoted to the study of multi-output learning and the development of novel approaches for addressing the challenges encountered. However, it lacks a comprehensive overview on different types of challenges of multi-output learning brought by the characteristics of the multiple outputs and the techniques proposed to overcome the challenges. This paper thus attempts to fill in this gap to provide a comprehensive review on this area. We first introduce different stages of the life cycle of the output labels. Then we present the paradigm on multi-output learning, including its myriads of output structures, definitions of its different sub-problems, model evaluation metrics and popular data repositories used in the study. Subsequently, we review a number of state-of-the-art multi-output learning methods, which are categorized based on the challenges.

Strong worst-case performance bounds for episodic reinforcement learning exist but fortunately in practice RL algorithms perform much better than such bounds would predict. Algorithms and theory that provide strong problem-dependent bounds could help illuminate the key features of what makes a RL problem hard and reduce the barrier to using RL algorithms in practice. As a step towards this we derive an algorithm for finite horizon discrete MDPs and associated analysis that both yields state-of-the art worst-case regret bounds in the dominant terms and yields substantially tighter bounds if the RL environment has small environmental norm, which is a function of the variance of the next-state value functions. An important benefit of our algorithmic is that it does not require apriori knowledge of a bound on the environmental norm. As a result of our analysis, we also help address an open learning theory question~\cite{jiang2018open} about episodic MDPs with a constant upper-bound on the sum of rewards, providing a regret bound with no $H$-dependence in the leading term that scales a polynomial function of the number of episodes.

Despite the great empirical success of deep reinforcement learning, its theoretical foundation is less well understood. In this work, we make the first attempt to theoretically understand the deep Q-network (DQN) algorithm (Mnih et al., 2015) from both algorithmic and statistical perspectives. In specific, we focus on a slight simplification of DQN that fully captures its key features. Under mild assumptions, we establish the algorithmic and statistical rates of convergence for the action-value functions of the iterative policy sequence obtained by DQN. In particular, the statistical error characterizes the bias and variance that arise from approximating the action-value function using deep neural network, while the algorithmic error converges to zero at a geometric rate. As a byproduct, our analysis provides justifications for the techniques of experience replay and target network, which are crucial to the empirical success of DQN. Furthermore, as a simple extension of DQN, we propose the Minimax-DQN algorithm for zero-sum Markov game with two players. Borrowing the analysis of DQN, we also quantify the difference between the policies obtained by Minimax-DQN and the Nash equilibrium of the Markov game in terms of both the algorithmic and statistical rates of convergence.

Many young professionals, who have started their journey into data science, and machine learning, face a common problem -- they have completed one or two basic online course, done some programming lessons, put up a couple of projects on Github, and then… then what? In one of my previous articles on Medium (published by the TDS Team), I discussed, at length, where you can find MOOC (Massive Open Online Course) for jump-starting your journey into data science and machine learning. That article assumed the reader to be a beginner and covers essential MOOCs, which are optimized for basic and intermediate learning. I wrote another detailed article specifically focused on the topic of mathematics concepts you need to master for data science and machine learning and which courses to study. Recently, I have been receiving a lot of messages in my personal email and LinkedIn inbox, mostly from bright, young professionals, asking similar questions and my suggestions about online courses. I mostly have a ready answer for those messages.

A common assumption in recommender systems (RS) is the existence of a best fixed recommendation strategy. Such strategy may be simple and work at the item level (e.g., in multi-armed bandit it is assumed one best fixed arm/item exists) or implement more sophisticated RS (e.g., the objective of A/B testing is to find the best fixed RS and execute it thereafter). We argue that this assumption is rarely verified in practice, as the recommendation process itself may impact the user’s preferences. For instance, a user may get bored by a strategy, while she may gain interest again, if enough time passed since the last time that strategy was used. In this case, a better approach consists in alternating different solutions at the right frequency to fully exploit their potential. In this paper, we first cast the problem as a Markov decision process, where the rewards are a linear function of the recent history of actions, and we show that a policy considering the long-term influence of the recommendations may outperform both fixed-action and contextual greedy policies. We then introduce an extension of the UCRL algorithm ( L IN UCRL ) to effectively balance exploration and exploitation in an unknown environment, and we derive a regret bound that is independent of the number of states. Finally, we empirically validate the model assumptions and the algorithm in a number of realistic scenarios.

While designing the state space of an MDP, it is common to include states that are transient or not reachable by any policy (e.g., in mountain car, the product space of speed and position contains configurations that are not physically reachable). This results in weakly-communicating or multi-chain MDPs. In this paper, we introduce TUCRL, the first algorithm able to perform efficient exploration-exploitation in any finite Markov Decision Process (MDP) without requiring any form of prior knowledge. In particular, for any MDP with $S^c$ communicating states, $A$ actions and $\Gamma^c \leq S^c$ possible communicating next states, we derive a $O(D^c \sqrt{\Gamma^c S^c A T}) regret bound, where $D^c$ is the diameter (i.e., the length of the longest shortest path between any two states) of the communicating part of the MDP. This is in contrast with optimistic algorithms (e.g., UCRL, Optimistic PSRL) that suffer linear regret in weakly-communicating MDPs, as well as posterior sampling or regularised algorithms (e.g., REGAL), which require prior knowledge on the bias span of the optimal policy to bias the exploration to achieve sub-linear regret. We also prove that in weakly-communicating MDPs, no algorithm can ever achieve a logarithmic growth of the regret without first suffering a linear regret for a number of steps that is exponential in the parameters of the MDP. Finally, we report numerical simulations supporting our theoretical findings and showing how TUCRL overcomes the limitations of the state-of-the-art.

A common assumption in recommender systems (RS) is the existence of a best fixed recommendation strategy. Such strategy may be simple and work at the item level (e.g., in multi-armed bandit it is assumed one best fixed arm/item exists) or implement more sophisticated RS (e.g., the objective of A/B testing is to find the best fixed RS and execute it thereafter). We argue that this assumption is rarely verified in practice, as the recommendation process itself may impact the user’s preferences. For instance, a user may get bored by a strategy, while she may gain interest again, if enough time passed since the last time that strategy was used. In this case, a better approach consists in alternating different solutions at the right frequency to fully exploit their potential. In this paper, we first cast the problem as a Markov decision process, where the rewards are a linear function of the recent history of actions, and we show that a policy considering the long-term influence of the recommendations may outperform both fixed-action and contextual greedy policies. We then introduce an extension of the UCRL algorithm ( L IN UCRL ) to effectively balance exploration and exploitation in an unknown environment, and we derive a regret bound that is independent of the number of states. Finally, we empirically validate the model assumptions and the algorithm in a number of realistic scenarios.