We present a reinforcement learning approach to goal based wealth management problems such as optimization of retirement plans or target dated funds. In such problems, an investor seeks to achieve a financial goal by making periodic investments in the portfolio while being employed, and periodically draws from the account when in retirement, in addition to the ability to re-balance the portfolio by selling and buying different assets (e.g. stocks). Instead of relying on a utility of consumption, we present G-Learner: a reinforcement learning algorithm that operates with explicitly defined one-step rewards, does not assume a data generation process, and is suitable for noisy data. Our approach is based on G-learning - a probabilistic extension of the Q-learning method of reinforcement learning. In this paper, we demonstrate how G-learning, when applied to a quadratic reward and Gaussian reference policy, gives an entropy-regulated Linear Quadratic Regulator (LQR). This critical insight provides a novel and computationally tractable tool for wealth management tasks which scales to high dimensional portfolios. In addition to the solution of the direct problem of G-learning, we also present a new algorithm, GIRL, that extends our goal-based G-learning approach to the setting of Inverse Reinforcement Learning (IRL) where rewards collected by the agent are not observed, and should instead be inferred. We demonstrate that GIRL can successfully learn the reward parameters of a G-Learner agent and thus imitate its behavior. Finally, we discuss potential applications of the G-Learner and GIRL algorithms for wealth management and robo-advising.

We develop a framework for the adaptive model predictive control of a linear system with unknown parameters and arbitrary bounded costs, in a critical setting where failures are costly and should be prevented at all time. Our approach builds on two ideas: first, we incorporate prior knowledge of the dynamics in the form of a known structure that shapes uncertainty, which can be tightened as we collect interaction data by building high-confidence regions through least-square regression. Second, in order to handle this uncertainty we formulate a robust control objective. Leveraging tools from the interval prediction literature, we convert the confidence regions on parameters into confidence sets on trajectories induced by the controls. These controls are then optimised resorting to tree-based planning methods. We eventually relax our modeling assumptions with a multi-model extension based on a data-driven robust model selection mechanism. The full procedure is designed to produce reasonable and safe behaviours at deployment while recovering an asymptotic optimality. We illustrate it on a practical case of autonomous driving at a crossroads intersection among vehicles with uncertain behaviours.

This paper presents the concept of an adaptive safe padding that forces Reinforcement Learning (RL) to synthesize optimal control policies while ensuring safety during the learning process. We express the safety requirements as a temporal logic formula. Enforcing the RL agent to stay safe during learning might limit the exploration in some safety-critical cases. However, we show that the proposed architecture is able to automatically handle the trade-off between efficient progress in exploration and ensuring strict safety. Theoretical guarantees are available on the convergence of the algorithm. Finally experimental results are provided to showcase the performance of the proposed method.

In recent years, Multifactorial Optimization (MFO) has gained a notable momentum in the research community. MFO is known for its inherent capability to efficiently address multiple optimization tasks at the same time, while transferring information among such tasks to improve their convergence speed. On the other hand, the quantum leap made by Deep Q Learning (DQL) in the Machine Learning field has allowed facing Reinforcement Learning (RL) problems of unprecedented complexity. Unfortunately, complex DQL models usually find it difficult to converge to optimal policies due to the lack of exploration or sparse rewards. In order to overcome these drawbacks, pre-trained models are widely harnessed via Transfer Learning, extrapolating knowledge acquired in a source task to the target task. Besides, meta-heuristic optimization has been shown to reduce the lack of exploration of DQL models. This work proposes a MFO framework capable of simultaneously evolving several DQL models towards solving interrelated RL tasks. Specifically, our proposed framework blends together the benefits of meta-heuristic optimization, Transfer Learning and DQL to automate the process of knowledge transfer and policy learning of distributed RL agents. A thorough experimentation is presented and discussed so as to assess the performance of the framework, its comparison to the traditional methodology for Transfer Learning in terms of convergence, speed and policy quality , and the intertask relationships found and exploited over the search process.

Multi-task reinforcement learning (RL) aims to simultaneously learn policies for solving many tasks. Several prior works have found that relabeling past experience with different reward functions can improve sample efficiency. Relabeling methods typically ask: if, in hindsight, we assume that our experience was optimal for some task, for what task was it optimal? In this paper, we show that hindsight relabeling is inverse RL, an observation that suggests that we can use inverse RL in tandem for RL algorithms to efficiently solve many tasks. We use this idea to generalize goal-relabeling techniques from prior work to arbitrary classes of tasks. Our experiments confirm that relabeling data using inverse RL accelerates learning in general multi-task settings, including goal-reaching, domains with discrete sets of rewards, and those with linear reward functions.

Practitioners often rely on compute-intensive domain randomization to ensure reinforcement learning policies trained in simulation can robustly transfer to the real world. Due to unmodeled nonlinearities in the real system, however, even such simulated policies can still fail to perform stably enough to acquire experience in real environments. In this paper we propose a novel method that guarantees a stable region of attraction for the output of a policy trained in simulation, even for highly nonlinear systems. Our core technique is to use "bias-shifted" neural networks for constructing the controller and training the network in the simulator. The modified neural networks not only capture the nonlinearities of the system but also provably preserve linearity in a certain region of the state space and thus can be tuned to resemble a linear quadratic regulator that is known to be stable for the real system. We have tested our new method by transferring simulated policies for a swing-up inverted pendulum to real systems and demonstrated its efficacy.

We study the problem of how to efficiently explore zero-sum games whose payoffs and dynamics are initially unknown. The agent is given a certain number of episodes to learn as much useful information about the game as possible. During this learning, the rewards obtained in the game are fictional and thus do not count toward the evaluation of the final strategy. After this exploration phase, the agent must recommend a strategy that should be minimally exploitable by an adversary (who has complete knowledge of the environment and can thus play optimally against it). This setup is called pure exploration in the single-agent reinforcement learning literature. This is an important problem for simulation-based games in which a black-box simulator is queried with strategies to obtain samples of the players' resulting utilities [33], as opposed to the rules of the game being explicitly given. For example, in many military settings, war game simulators are used to generate strategies, and then the strategies need to be ready to deploy in case of actual war [17]. Another prevalent example is finance, where trading strategies are generated in simulation, and then they need to be ready for live trading. A third example is video games such as Dota 2 [4] and Starcraft II [31], where AIs can be trained largely through self-play.

In this paper, we propose a derivative-free model learning framework for Reinforcement Learning (RL) algorithms based on Gaussian Process Regression (GPR). In many mechanical systems, only positions can be measured by the sensing instruments. Then, instead of representing the system state as suggested by the physics with a collection of positions, velocities, and accelerations, we define the state as the set of past position measurements. However, the equation of motions derived by physical first principles cannot be directly applied in this framework, being functions of velocities and accelerations. For this reason, we introduce a novel derivative-free physically-inspired kernel, which can be easily combined with nonparametric derivative-free Gaussian Process models. Tests performed on two real platforms show that the considered state definition combined with the proposed model improves estimation performance and data-efficiency w.r.t. traditional models based on GPR. Finally, we validate the proposed framework by solving two RL control problems for two real robotic systems.

We consider the problem of finding Nash equilibrium for two-player turn-based zero-sum games. Inspired by the AlphaGo Zero (AGZ) algorithm, we develop a Reinforcement Learning based approach. Specifically, we propose Explore-Improve-Supervise (EIS) method that combines "exploration", "policy improvement"' and "supervised learning" to find the value function and policy associated with Nash equilibrium. We identify sufficient conditions for convergence and correctness for such an approach. For a concrete instance of EIS where random policy is used for "exploration", Monte-Carlo Tree Search is used for "policy improvement" and Nearest Neighbors is used for "supervised learning", we establish that this method finds an $\varepsilon$-approximate value function of Nash equilibrium in $\widetilde{O}(\varepsilon^{-(d+4)})$ steps when the underlying state-space of the game is continuous and $d$-dimensional. This is nearly optimal as we establish a lower bound of $\widetilde{\Omega}(\varepsilon^{-(d+2)})$ for any policy.

It has been a trend in the Reinforcement Learning literature to derive sample complexity bounds: a bound on how many experiences with the environment are required to obtain an $\varepsilon$-optimal policy. In the discounted cost, infinite horizon setting, all of the known bounds have a factor that is a polynomial in $1/(1-\beta)$, where $\beta < 1$ is the discount factor. For a large discount factor, these bounds seem to imply that a very large number of samples is required to achieve an $\varepsilon$-optimal policy. The objective of the present work is to introduce a new class of algorithms that have sample complexity uniformly bounded for all $\beta < 1$. One may argue that this is impossible, due to a recent min-max lower bound. The explanation is that this previous lower bound is for a specific problem, which we modify, without compromising the ultimate objective of obtaining an $\varepsilon$-optimal policy. Specifically, we show that the asymptotic variance of the Q-learning algorithm, with an optimized step-size sequence, is a quadratic function of $1/(1-\beta)$; an expected, and essentially known result. The new relative Q-learning algorithm proposed here is shown to have asymptotic variance that is a quadratic in $1/(1- \rho \beta)$, where $1 - \rho > 0$ is the spectral gap of an optimal transition matrix.