Vacant taxi drivers' passenger seeking process in a road network generates additional vehicle miles traveled, adding congestion and pollution into the road network and the environment. This paper aims to employ a Markov Decision Process (MDP) to model idle e-hailing drivers' optimal sequential decisions in passenger-seeking. Transportation network companies (TNC) or e-hailing (e.g., Didi, Uber) drivers exhibit different behaviors from traditional taxi drivers because e-hailing drivers do not need to actually search for passengers. Instead, they reposition themselves so that the matching platform can match a passenger. Accordingly, we incorporate e-hailing drivers' new features into our MDP model. We then use 44,160 Didi drivers' 3-day trajectories to train the model. To validate the effectiveness of the model, a Monte Carlo simulation is conducted to simulate the performance of drivers under the guidance of the optimal policy, which is then compared with the performance of drivers following one baseline heuristic, namely, the local hotspot strategy. The results show that our model is able to achieve a 26% improvement over the local hotspot strategy in terms of the rate of return. The proposed MDP model captures the supply-demand ratio considering the fact that the number of drivers in this study is sufficiently large and thus the number of unmatched orders is assumed to be negligible. To better incorporate the competition among multiple drivers into the model, we have also devised and calibrated a dynamic adjustment strategy of the order matching probability.

We derive and analyze learning algorithms for policy evaluation, apprenticeship learning, and policy gradient for average reward criteria. Existing algorithms explicitly require an upper bound on the mixing time. In contrast, we build on ideas from Markov chain theory and derive sampling algorithms that do not require such an upper bound. For these algorithms, we provide theoretical bounds on their sample-complexity and running time.

For this For reinforcement learning in MDP with changes in reward problem setting, we propose an algorithm and function and transition probabilities, we provide provide performance guarantees for the regret an algorithm, UCRL with Restarts, a version of UCRL evaluated against the optimal non-stationary [Jaksch et al., 2010], which restarts according to a schedule policy. The upper bound on the regret is given dependent on the variation in the MDP (defined in terms of the total variation in the MDP. in Section 2 below). We derive a high-probability upper This is the first variational regret bound for the bound on the cumulative regret of our algorithm of general reinforcement learning setting.

Microsoft's latest breakthrough could make DNA-based data centers possible

Recent advances in deep reinforcement learning (RL) have shown remarkable success on challenging tasks (Andrychowicz et al., 2018; Mnih et al., 2015; Silver et al., 2016). However, data efficiency and robustness to new contexts remain persistent challenges for deep RL algorithms, especially when the goal is for agents to learn practical tasks with limited supervision. Drawing inspiration from self-supervised "play" in human development (Gopnik et al., 1999; Settles, 2011), we introduce an agent that learns object-centric representations of its environment without supervision and subsequently harnesses these to learn policies efficiency and robustly. Our agent, which we call Curious Object-Based seaRch Agent (COBRA), brings together three key ingredients: (i) learning representations of the world in terms of objects, (ii) curiosity-driven exploration, and (iii) model based RL. The benefits of this synthesis are data efficiency and policy robustness. To put this into practice, we introduce the following technical contributions: - A method for learning action-conditioned dynamics over slot-structured object-centric representations that requires no supervision and is trained from raw pixels.

Reading Comprehension has received significant attention in recent years as high quality Question Answering (QA) datasets have become available. Despite state-of-the-art methods achieving strong overall accuracy, Multi-Hop (MH) reasoning remains particularly challenging. To address MH-QA specifically, we propose a Deep Reinforcement Learning based method capable of learning sequential reasoning across large collections of documents so as to pass a query-aware, fixed-size context subset to existing models for answer extraction. Our method is comprised of two stages: a linker, which decomposes the provided support documents into a graph of sentences, and an extractor, which learns where to look based on the current question and already-visited sentences. The result of the linker is a novel graph structure at the sentence level that preserves logical flow while still allowing rapid movement between documents. Importantly, we demonstrate that the sparsity of the resultant graph is invariant to context size. This translates to fewer decisions required from the Deep-RL trained extractor, allowing the system to scale effectively to large collections of documents. The importance of sequential decision making in the document traversal step is demonstrated by comparison to standard IE methods, and we additionally introduce a BM25-based IR baseline that retrieves documents relevant to the query only. We examine the integration of our method with existing models on the recently proposed QAngaroo benchmark and achieve consistent increases in accuracy across the board, as well as a 2-3x reduction in training time.

Classically, imitation learning algorithms have been developed for idealized situations, e.g., the demonstrations are often required to be collected in the exact same environment and usually include the demonstrator's actions. Recently, however, the research community has begun to address some of these shortcomings by offering algorithmic solutions that enable imitation learning from observation (IfO), e.g., learning to perform a task from visual demonstrations that may be in a different environment and do not include actions. Motivated by the fact that agents often also have access to their own internal states (i.e., proprioception), we propose and study an IfO algorithm that leverages this information in the policy learning process. The proposed architecture learns policies over proprioceptive state representations and compares the resulting trajectories visually to the demonstration data. We experimentally test the proposed technique on several MuJoCo domains and show that it outperforms other imitation from observation algorithms by a large margin.

Phishing is the simplest form of cybercrime with the objective of baiting people into giving away delicate information such as individually recognizable data, banking and credit card details, or even credentials and passwords. This type of simple yet most effective cyber-attack is usually launched through emails, phone calls, or instant messages. The credential or private data stolen are then used to get access to critical records of the victims and can result in extensive fraud and monetary loss. Hence, sending malicious messages to victims is a stepping stone of the phishing procedure. A phisher usually setups a deceptive website, where the victims are conned into entering credentials and sensitive information. It is therefore important to detect these types of malicious websites before causing any harmful damages to victims. Inspired by the evolving nature of the phishing websites, this paper introduces a novel approach based on deep reinforcement learning to model and detect malicious URLs. The proposed model is capable of adapting to the dynamic behavior of the phishing websites and thus learn the features associated with phishing website detection. I. INTRODUCTION Phishing is a form of cyber attack typically performed by sending false correspondences that seem to be originated from a legitimate source.

Here we propose a method to extremely accelerate NAS, without reinforcement learning or gradient, just by sampling architectures from a distribution and comparing these architectures, estimating their relative performance rather than absolute performance, iteratively updating parameters of the distribution while training. Search codes will be released by Sherwood later!

Inverse reinforcement learning (IRL) is an ill-posed inverse problem since expert demonstrations may infer many solutions of reward functions which is hard to recover by local search methods such as a gradient method. In this paper, we generalize the original IRL problem to recover a probability distribution for reward functions. We call such a generalized problem stochastic inverse reinforcement learning (SIRL) which is first formulated as an expectation optimization problem. We adopt the Monte Carlo expectation-maximization (MCEM) method, a global search method, to estimate the parameter of the probability distribution as the first solution to SIRL. With our approach, it is possible to observe the deep intrinsic property in IRL from a global viewpoint, and the technique achieves a considerable robust recovery performance on the classic learning environment, objectworld.