Learning based supervisor synthesis of POMDP for PCTL specifications

The ability to reason about and respond to their own emotional states can enhance the believability of Non-Player Characters (NPCs). In this paper, we use a Partially Observable Markov Decision Process (POMDP)-based framework to model emotion over time. A two-level appraisal model, involving quick and reactive vs. slow and deliberate appraisals, is proposed for the creation of affective autonomous characters based on POMDPs, wherein the probability of goal satisfaction is used in an appraisal and reappraisal process for emotion generation. We not only extend Probabilistic Computation Tree Logic (PCTL) for reasoning about the properties of emotional states based on POMDPs but also illustrate how four reactive (primary) emotions and nine deliberate (secondary) emotions can be derived by combining PCTL with the belief-desire theory of emotion. The results of an empirical study suggest that the proposed model can be used to create characters that appear to be more believable and more intelligent.

With the advances of Artificial Intelligence (AI) and autonomous robotics, robots are increasingly involved closely with human society, such as in manufacturing system, autonomous driving and intelligent service robots. While robots are good at handling repeated routine work, human are more adaptive and flexible to changing factors that may bring uncertainties and cost non-trivial efforts for robots to overcome. Thus efficient collaboration between human and robots can bring huge economic profits, but also require high standard for robot controller during task completion because the safety of human must be guaranteed. In this paper, we build a formal design theory enabling an automated human-robot collaboration with performance guarantees. Markov Decision Processes (MDPs) are used to model robots with uncertainties and Partially Observable Markov Decision Processes (POMDPs) are used to model human with hidden intents. Within a supervisory control framework, we propose to use L* learning algorithm to learn a correct supervisor such that global collaboration task given as Probabilistic Computation Tree Logic (PCTL) can be satisfied. An example from manufacturing system is discussed to further illustrate the design method.

The framework of partially observable Markov decision processes (POMDPs) offers a standard approach to model uncertainty in many robot tasks. Traditionally, POMDPs are formulated with optimality objectives. In this article, we study a different formulation of POMDPs with <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Boolean objectives</i> . For robotic domains that require a correctness guarantee of accomplishing tasks, Boolean objectives are natural formulations. We investigate the problem of POMDPs with a common Boolean objective: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">safe reachability</i> , requiring that the robot eventually reaches a goal state with a probability above a threshold while keeping the probability of visiting unsafe states below a different threshold. Our approach builds upon the previous work that represents POMDPs with Boolean objectives using symbolic constraints. We employ a satisfiability modulo theories (SMTs) solver to efficiently search for solutions, i.e., policies or conditional plans that specify the action to take contingent on every possible event. A full policy or conditional plan is generally expensive to compute. To improve computational efficiency, we introduce the notion of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">partial conditional plans</i> that cover sampled events to approximate a full conditional plan. Our approach constructs a partial conditional plan parameterized by a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">replanning probability</i> . We prove that the failure rate of the constructed partial conditional plan is bounded by the replanning probability. Our approach allows users to specify an appropriate bound on the replanning probability to balance efficiency and correctness. Moreover, we update this bound properly to quickly detect whether the current partial conditional plan meets the bound and avoid unnecessary computation. In addition, to further improve the efficiency, we cache partial conditional plans for sampled belief states and reuse these cached plans if possible. We validate our approach in several robotic domains. The results show that our approach outperforms a previous policy synthesis approach for POMDPs with safe-reachability objectives in these domains. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This article was motivated by two observations. On the one hand, in robotics applications where uncertainty in sensing and actions is present, the solution to the classical partially observable Markov decision process (POMDP) formulation is expensive to compute in general. On the other hand, in certain practical scenarios, formulations other than the classical POMDP make a lot of sense and can provide flexibility in balancing efficiency and correctness. This article considers a modified POMDP formulation that includes a Boolean objective, namely safe reachability. This article uses the notion of a partial conditional plan. Rather than explicitly enumerating all possible observations to construct a full conditional plan, this work samples a subset of all observations to ensure bounded replanning probability. Our theoretical and empirical results show that the failure rate of the constructed partial conditional plan is bounded by the replanning probability. Moreover, these partial conditional plans can be cached to further improve the performance. Our results suggest that for domains where replanning is easy, increasing the replanning probability bound usually leads to better scalability, and for domains where replanning is difficult or impossible in some states, we can decrease the bound and allocate more computation time to achieve a higher success rate. Hence, in certain cases, the practitioner can take advantage of their knowledge of the problem domain to scale to larger problems. Preliminary physical experiments suggest that this approach is applicable to real-world robotic domains, but it requires a discrete representation of the workspace. How to deal with continuous workspace directly is an interesting future direction.

Aiming to reduce energy consumption, minimize transmission latency, and optimize the long-term performance of Wireless Sensor Networks (WSN), this paper proposes an improved routing model of Directed Diffusion. We formulated the routing issue as a partially observable Markov decision process (POMDP), such that the optimal nodes selection policy yielded by maximizing the expected reward. Since the solution of POMDP is ready-made, the attention is concentrated only on the formulation of the routing issue into a POMDP model, more specifically, the definition of the tuple which contains states, actions, observations, transition probabilities, observation probabilities, and reward function. Our experimental study shows that the POMDP-based Directed Diffusion routing outperforms other routing techniques in terms of three widely used metrics including average dissipated energy, average delay, and average lifetime.

In this article, we consider a subclass of partially observable Markov decision process (POMDP) problems which we termed confounding POMDPs. In these types of POMDPs, temporal difference (TD)-based reinforcement learning (RL) algorithms struggle, as TD error cannot be easily derived from observations. We solve these types of problems using a new bio-inspired neural architecture that combines a modulated Hebbian network (MOHN) with deep Q-network (DQN), which we call modulated Hebbian plus Q-network architecture (MOHQA). The key idea is to use a Hebbian network with rarely correlated bio-inspired neural traces to bridge temporal delays between actions and rewards when confounding observations and sparse rewards result in inaccurate TD errors. In MOHQA, DQN learns low-level features and control, while the MOHN contributes to high-level decisions by associating rewards with past states and actions. Thus, the proposed architecture combines two modules with significantly different learning algorithms, a Hebbian associative network and a classical DQN pipeline, exploiting the advantages of both. Simulations on a set of POMDPs and on the Malmo environment show that the proposed algorithm improved DQN's results and even outperformed control tests with advantage-actor critic (A2C), quantile regression DQN with long short-term memory (QRDQN + LSTM), Monte Carlo policy gradient (REINFORCE), and aggregated memory for reinforcement learning (AMRL) algorithms on most difficult POMDPs with confounding stimuli and sparse rewards.

&lt;p&gt;Partially Observable Markov Decision Processes (POMDPs) is a modeling framework that allow decision makers to make decisions in uncertain environments. This framework is particularly suited for sequential decision making problems for which the uncertainties are revealed in stages. Constrained POMDPs (CPOMDPs) extend this framework by also accounting for various restrictions in the environment. This PhD thesis focuses on POMDPs, CPOMDPs and its extensions, proposes novel solution methods and studies applications of POMDPs in disease screening. First, we focus on developing efficient versions of the well-known algorithms for POMDPs. POMDPs are notoriously difficult to solve optimally, and exact solution algorithms typically fail to solve the instances with more than a few core states. In this regard, we consider scalable algorithms solve the POMDP problems. Specifically, we propose distributed and parallel versions of the exact solution methods such as Monahan's algorithm and the incremental pruning algorithm as well as the heuristics such as Lovejoy's upper bound and lower bound methods. Next, we study the CPOMDPs and propose linear programming-based solution methods as a flexible approach for those. We perform a detailed numerical study with five different problem instances to showcase the viability of the proposed solution approach for CPOMDPs. Lastly, we propose multi-objective CPOMDP models for the breast cancer screening problem. These CPOMDP models extend the previously proposed POMDP models for breast cancer screening by considering different screening modalities and simultaneously optimizing for expected total quality-adjusted life years maximization and life-time risk minimization objectives.&lt;/p&gt;

Partially observable Markov decision processes (POMDPs) are models for sequential decision-making under uncertainty and incomplete information. Machine learning methods typically train recurrent neural networks (RNN) as effective representations of POMDP policies that can efficiently process sequential data. However, it is hard to verify whether the POMDP driven by such RNN-based policies satisfies safety constraints, for instance, given by temporal logic specifications. We propose a novel method that combines techniques from machine learning with the field of formal methods: training an RNN-based policy and then automatically extracting a so-called finite-state controller (FSC) from the RNN. Such FSCs offer a convenient way to verify temporal logic constraints. Implemented on a POMDP, they induce a Markov chain, and probabilistic verification methods can efficiently check whether this induced Markov chain satisfies a temporal logic specification. Using such methods, if the Markov chain does not satisfy the specification, a byproduct of verification is diagnostic information about the states in the POMDP that are critical for the specification. The method exploits this diagnostic information to either adjust the complexity of the extracted FSC or improve the policy by performing focused retraining of the RNN. The method synthesizes policies that satisfy temporal logic specifications for POMDPs with up to millions of states, which are three orders of magnitude larger than comparable approaches.

Partially Observable Markov Decision Processes (POMDP) provides piecewise-linear a natural and principled framework for sequential decision-making under uncertainty. However, large-scale POMDP suffers from the exponential growth of the belief points and policy trees space. We present a new point-based incremental pruning algorithm based on the piecewise linearity and convexity of the value function. Instead of reasoning about the whole belief space when pruning the cross-sums in POMDP policy construction, our algorithm uses belief points to perform approximate pruning by generating policy trees, and get the optimal policy in real-time belief states. The empirical results indicate that point-based incremental pruning for heuristic search methods can handle large POMDP domains efficiently.

We consider the general problem of optimal stochastic control of a dedicated-platform that processes one primary function or task (target-task). The dedicated-platform has two modes of action at each period of time: it can attempt to process the target-task at the given period of time, or suspend the target-task for later completion. We formulate the optimal trade-off between the processing cost and the latency in completion of the target-task as a Partially Observable Markov Decision Process (POMDP). By reformulating this POMDP as a Markovian search problem, we prove that the optimal control policies are threshold in nature. Threshold policies are computationally efficient and inexpensive to implement in real time systems. Numerical results demonstrate the effectiveness of these threshold based operating algorithms as compared to non-optimal heuristic algorithms.KeywordsPartially Observable Markov Decision Process (POMDP)optimal threshold policiesdynamic programmingBellman equationtwo-state Markov chainoptimal searchoverlook probabilitysufficient statisticsHidden Markov Model (HMM)

Sepsis is a life-threatening condition, caused by the body’s extreme response to an infection. In the United States, 1.7 million cases of sepsis occur annually, resulting in 265,000 deaths. Delayed diagnosis and treatment are associated with higher mortality rates. An exponential rise in the availability of medical data has allowed for the development of sophisticated machine learning algorithms to predict sepsis earlier than the onset. However, these models often underperform, as the training data are retrospective and do not fully capture the uncertain future. In this study, we develop a novel framework, which we refer to as MLePOMDP, to leverage and combine the underlying, high-level knowledge about sepsis progression and machine learning (ML) for classification. Specifically, we use a hidden Markov model to describe sepsis development at a high level, where the ML model makes the higher-order “observations” from temporal data. Consequently, a partially observable Markov decision process (POMDP) model is developed to make classification decisions. We analytically establish that the optimal policy is of threshold-type, which we exploit to efficiently optimize MLePOMDP. MLePOMDP is calibrated and tested using high-frequency physiological data collected from bedside monitors. Different from past POMDP-based frameworks, MLePOMDP is developed for a prediction task using a very small state definition, produces highly interpretable results, and accounts for a novel and clinically meaningful action space. Our results show that MLePOMDP outperforms machine learning–based benchmarks by up to 8% in precision. Importantly, MLePOMDP is able to reduce false alarms by up to 28%. An additional experiment is conducted to show the generalizability of MLePOMDP to different patient cohorts. Summary of Contribution: This study develops a novel real-time decision support framework for early sepsis prediction by integrating well-known machine learning models (random forest and neural networks) with a well-established sequential decision-making model, namely, a partially observable Markov decision process (POMDP). The structural properties of the optimal policy are further explored and a threshold-type structure is established, which is then leveraged to develop a customized algorithm to solve the problem more efficiently. The resulting framework demonstrates the benefit of applying POMDPs to augment machine learning outputs. Specifically, the framework results in the reduction of false alarms in sepsis predictions where decisions are made in real time, hence improving the overall prediction precision.

Ambiguous partially observable Markov decision processes: Structural results and applications

Optimal policies for partially observed Markov decision processes (POMDPs) are history-dependent: Decisions are made based on the entire history of observations. Memoryless policies, which take decisions based on the last observation only, are generally considered useless in the literature because we can construct POMDP instances for which optimal memoryless policies are arbitrarily worse than history-dependent ones. Our purpose is to challenge this belief. We show that optimal memoryless policies can be computed efficiently using mixed integer linear programming (MILP), and perform reasonably well on a wide range of instances from the literature. When strengthened with valid inequalities, the linear relaxation of this MILP provides high quality upper-bounds on the value of an optimal history dependent policy. Furthermore, when used with a finite horizon POMDP problem with memoryless policies as rolling optimization problem, a model predictive control approach leads to an efficient history-dependent policy, which we call the short memory in the future (SMF) policy. Basically, the SMF policy leverages these memoryless policies to build an approximation of the Bellman value function. Numerical experiments show the efficiency of our approach on benchmark instances from the literature.

Resolving structural uncertainty in natural resources management using POMDP approaches

In this article, we study the problem of controlling a partially observed Markov decision process (POMDP) to either aid or hinder the estimation of its state trajectory. We encode the estimation objectives via the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">smoother entropy</i> , which is the conditional entropy of the state trajectory given measurements and controls. Consideration of the smoother entropy contrasts with previous approaches that instead resort to marginal (or instantaneous) state entropies due to tractability concerns. By establishing novel expressions for the smoother entropy in terms of the POMDP belief state, we show that both the problems of minimizing and maximizing the smoother entropy in POMDPs can surprisingly be reformulated as belief-state Markov decision processes with concave cost and value functions. The significance of these reformulations is that they render the smoother entropy a tractable optimization objective, with structural properties amenable to the use of standard POMDP solution techniques for <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">both</i> active estimation and obfuscation. Simulations illustrate that optimization of the smoother entropy leads to superior trajectory estimation and obfuscation compared to alternative approaches.

Partially Observable Markov Decision Processes (POMDPs) are a powerful framework for planning under uncertainty. They allow to model state uncertainty as a belief probability distribution. Approximate solvers based on Monte Carlo sampling show great success to relax the computational demand and perform online planning. However, scaling to complex realistic domains with many actions and long planning horizons is still a major challenge, and a key point to achieve good performance is guiding the action-selection process with domain-dependent policy heuristics which are tailored for the specific application domain. We propose to learn high-quality heuristics from POMDP traces of executions generated by any solver. We convert the belief-action pairs to a logical semantics, and exploit data- and time-efficient Inductive Logic Programming (ILP) to generate interpretable belief-based policy specifications, which are then used as online heuristics. We evaluate thoroughly our methodology on two notoriously challenging POMDP problems, involving large action spaces and long planning horizons, namely, rocksample and pocman. Considering different state-of-the-art online POMDP solvers, including POMCP, DESPOT and AdaOPS, we show that learned heuristics expressed in Answer Set Programming (ASP) yield performance superior to neural networks and similar to optimal handcrafted task-specific heuristics within lower computational time. Moreover, they well generalize to more challenging scenarios not experienced in the training phase (e.g., increasing rocks and grid size in rocksample, incrementing the size of the map and the aggressivity of ghosts in pocman).