In recent years, machine learning algorithms, and more specially, deep learning algorithms, have been widely used in many fields, including cyber security. However, machine learning systems are vulnerable to adversarial attacks, and this limits the application of machine learning, especially in non-stationary, adversarial environments, such as the cyber security domain, where actual adversaries (e.g., malware developers) exist. This paper comprehensively summarizes the latest research on adversarial attacks against security solutions that are based on machine learning techniques and presents the risks they pose to cyber security solutions. First, we discuss the unique challenges of implementing end-to-end adversarial attacks in the cyber security domain. Following that, we define a unified taxonomy, where the adversarial attack methods are characterized based on their stage of occurrence, and the attacker's goals and capabilities. Then, we categorize the applications of adversarial attack techniques in the cyber security domain. Finally, we use our taxonomy to shed light on gaps in the cyber security domain that have already been addressed in other adversarial learning domains and discuss their impact on future adversarial learning trends in the cyber security domain.

My aim, as always, was to keep the projects as diverse as possible so you can pick the ones that fit into your data science journey. If you're a beginner, I would suggest starting with the PalmerPenguins dataset as most folks aren't even aware of it right now. A great chance to get a head start. I would love to hear your thoughts on which open source project you found the most useful. Or let me know if you want me to feature any other data science projects here or in next month's edition.

Specifying a Reinforcement Learning (RL) task involves choosing a suitable planning horizon, which is typically modeled by a discount factor. It is known that applying RL algorithms with a lower discount factor can act as a regularizer, improving performance in the limited data regime. Yet the exact nature of this regularizer has not been investigated. In this work, we fill in this gap. For several Temporal-Difference (TD) learning methods, we show an explicit equivalence between using a reduced discount factor and adding an explicit regularization term to the algorithm's loss. Motivated by the equivalence, we empirically study this technique compared to standard $L_2$ regularization by extensive experiments in discrete and continuous domains, using tabular and functional representations. Our experiments suggest the regularization effectiveness is strongly related to properties of the available data, such as size, distribution, and mixing rate.

In recent years, reinforcement learning (RL) systems with general goals beyond a cumulative sum of rewards have gained traction, such as in constrained problems, exploration, and acting upon prior experiences. In this paper, we consider policy optimization in Markov Decision Problems, where the objective is a general concave utility function of the state-action occupancy measure, which subsumes several of the aforementioned examples as special cases. Such generality invalidates the Bellman equation. As this means that dynamic programming no longer works, we focus on direct policy search. Analogously to the Policy Gradient Theorem \cite{sutton2000policy} available for RL with cumulative rewards, we derive a new Variational Policy Gradient Theorem for RL with general utilities, which establishes that the parametrized policy gradient may be obtained as the solution of a stochastic saddle point problem involving the Fenchel dual of the utility function. We develop a variational Monte Carlo gradient estimation algorithm to compute the policy gradient based on sample paths. We prove that the variational policy gradient scheme converges globally to the optimal policy for the general objective, though the optimization problem is nonconvex. We also establish its rate of convergence of the order $O(1/t)$ by exploiting the hidden convexity of the problem, and proves that it converges exponentially when the problem admits hidden strong convexity. Our analysis applies to the standard RL problem with cumulative rewards as a special case, in which case our result improves the available convergence rate.

Off-policy evaluation (OPE) is the problem of evaluating new policies using historical data obtained from a different policy. Off-policy learning (OPL), on the other hand, is the problem of finding an optimal policy using historical data. In recent OPE and OPL contexts, most of the studies have focused on one-player cases, and not on more than two-player cases. In this study, we propose methods for OPE and OPL in two-player zero-sum Markov games. For OPE, we estimate exploitability that is often used as a metric for determining how close a strategy profile is to a Nash equilibrium in two-player zero-sum games. For OPL, we calculate maximin policies as Nash equilibrium strategies over the historical data. We prove the exploitability estimation error bounds for OPE and regret bounds for OPL based on the doubly robust and double reinforcement learning estimators. Finally, we demonstrate the effectiveness and performance of the proposed methods through experiments.

The question of how to determine which states and actions are responsible for a certain outcome is known as the credit assignment problem and remains a central research question in reinforcement learning and artificial intelligence. Eligibility traces enable efficient credit assignment to the recent sequence of states and actions experienced by the agent, but not to counterfactual sequences that could also have led to the current state. In this work, we introduce expected eligibility traces. Expected traces allow, with a single update, to update states and actions that could have preceded the current state, even if they did not do so on this occasion. We discuss when expected traces provide benefits over classic (instantaneous) traces in temporal-difference learning, and show that sometimes substantial improvements can be attained. We provide a way to smoothly interpolate between instantaneous and expected traces by a mechanism similar to bootstrapping, which ensures that the resulting algorithm is a strict generalisation of TD($\lambda$). Finally, we discuss possible extensions and connections to related ideas, such as successor features.

In cooperative multi-agent reinforcement learning, a collection of agents learns to interact in a shared environment to achieve a common goal. We propose the use of reward machines (RM) -- Mealy machines used as structured representations of reward functions -- to encode the team's task. The proposed novel interpretation of RMs in the multi-agent setting explicitly encodes required teammate interdependencies and independencies, allowing the team-level task to be decomposed into sub-tasks for individual agents. We define such a notion of RM decomposition and present algorithmically verifiable conditions guaranteeing that distributed completion of the sub-tasks leads to team behavior accomplishing the original task. This framework for task decomposition provides a natural approach to decentralized learning: agents may learn to accomplish their sub-tasks while observing only their local state and abstracted representations of their teammates. We accordingly propose a decentralized q-learning algorithm. Furthermore, in the case of undiscounted rewards, we use local value functions to derive lower and upper bounds for the global value function corresponding to the team task. Experimental results in three discrete settings exemplify the effectiveness of the proposed RM decomposition approach, which converges to a successful team policy two orders of magnitude faster than a centralized learner and significantly outperforms hierarchical and independent q-learning approaches.

Due to the realization that deep reinforcement learning algorithms trained on high-dimensional tasks can strongly overfit to their training environments, there have been several studies that investigated the generalization performance of these algorithms. However, there has been no similar study that evaluated the generalization performance of algorithms that were specifically designed for generalization, i.e. meta-reinforcement learning algorithms. In this paper, we assess the generalization performance of these algorithms by leveraging high-dimensional, procedurally generated environments. We find that these algorithms can display strong overfitting when they are evaluated on challenging tasks. We also observe that scalability to high-dimensional tasks with sparse rewards remains a significant problem among many of the current meta-reinforcement learning algorithms. With these results, we highlight the need for developing meta-reinforcement learning algorithms that can both generalize and scale.

While traditionally a labour intensive task, the testing of game content is progressively becoming more automated. Among the many directions in which this automation is taking shape, automatic play-testing is one of the most promising thanks also to advancements of many supervised and reinforcement learning (RL) algorithms. However these type of algorithms, while extremely powerful, often suffer in production environments due to issues with reliability and transparency in their training and usage. In this research work we are investigating and evaluating strategies to apply the popular RL method Proximal Policy Optimization (PPO) in a casual mobile puzzle game with a specific focus on improving its reliability in training and generalization during game playing. We have implemented and tested a number of different strategies against a real-world mobile puzzle game (Lily's Garden from Tactile Games). We isolated the conditions that lead to a failure in either training or generalization during testing and we identified a few strategies to ensure a more stable behaviour of the algorithm in this game genre.

Model-based reinforcement learning approaches leverage a forward dynamics model to support planning and decision making, which, however, may fail catastrophically if the model is inaccurate. Although there are several existing methods dedicated to combating the model error, the potential of the single forward model is still limited. In this paper, we propose to additionally construct a backward dynamics model to reduce the reliance on accuracy in forward model predictions. We develop a novel method, called Bidirectional Model-based Policy Optimization (BMPO) to utilize both the forward model and backward model to generate short branched rollouts for policy optimization. Furthermore, we theoretically derive a tighter bound of return discrepancy, which shows the superiority of BMPO against the one using merely the forward model. Extensive experiments demonstrate that BMPO outperforms state-of-the-art model-based methods in terms of sample efficiency and asymptotic performance.