Compositional Learning of Dynamical System Models Using Port-Hamiltonian Neural Networks

Many dynamical systems -- from robots interacting with their surroundings to large-scale multiphysics systems -- involve a number of interacting subsystems. Toward the objective of learning composite models of such systems from data, we present i) a framework for compositional neural networks, ii) algorithms to train these models, iii) a method to compose the learned models, iv) theoretical results that bound the error of the resulting composite models, and v) a method to learn the composition itself, when it is not known a prior. The end result is a modular approach to learning: neural network submodels are trained on trajectory data generated by relatively simple subsystems, and the dynamics of more complex composite systems are then predicted without requiring additional data generated by the composite systems themselves. We achieve this compositionality by representing the system of interest, as well as each of its subsystems, as a port-Hamiltonian neural network (PHNN) -- a class of neural ordinary differential equations that uses the port-Hamiltonian systems formulation as inductive bias. We compose collections of PHNNs by using the system's physics-informed interconnection structure, which may be known a priori, or may itself be learned from data. We demonstrate the novel capabilities of the proposed framework through numerical examples involving interacting spring-mass-damper systems. Models of these systems, which include nonlinear energy dissipation and control inputs, are learned independently. Accurate compositions are learned using an amount of training data that is negligible in comparison with that required to train a new model from scratch. Finally, we observe that the composite PHNNs enjoy properties of port-Hamiltonian systems, such as cyclo-passivity -- a property that is useful for control purposes.

Learning Interpretable Temporal Properties from Positive Examples Only

We consider the problem of explaining the temporal behavior of black-box systems using human-interpretable models. To this end, based on recent research trends, we rely on the fundamental yet interpretable models of deterministic finite automata (DFAs) and linear temporal logic (LTL) formulas. In contrast to most existing works for learning DFAs and LTL formulas, we rely on only positive examples. Our motivation is that negative examples are generally difficult to observe, in particular, from black-box systems. To learn meaningful models from positive examples only, we design algorithms that rely on conciseness and language minimality of models as regularizers. To this end, our algorithms adopt two approaches: a symbolic and a counterexample-guided one. While the symbolic approach exploits an efficient encoding of language minimality as a constraint satisfaction problem, the counterexample-guided one relies on generating suitable negative examples to prune the search. Both the approaches provide us with effective algorithms with theoretical guarantees on the learned models. To assess the effectiveness of our algorithms, we evaluate all of them on synthetic data.

Categorical semantics of compositional reinforcement learning

Reinforcement learning (RL) often requires decomposing a problem into subtasks and composing learned behaviors on these tasks. Compositionality in RL has the potential to create modular subtask units that interface with other system capabilities. However, generating compositional models requires the characterization of minimal assumptions for the robustness of the compositional feature. We develop a framework for a \emph{compositional theory} of RL using a categorical point of view. Given the categorical representation of compositionality, we investigate sufficient conditions under which learning-by-parts results in the same optimal policy as learning on the whole. In particular, our approach introduces a category $\mathsf{MDP}$, whose objects are Markov decision processes (MDPs) acting as models of tasks. We show that $\mathsf{MDP}$ admits natural compositional operations, such as certain fiber products and pushouts. These operations make explicit compositional phenomena in RL and unify existing constructions, such as puncturing hazardous states in composite MDPs and incorporating state-action symmetry. We also model sequential task completion by introducing the language of zig-zag diagrams that is an immediate application of the pushout operation in $\mathsf{MDP}$.

Deceptive Planning for Resource Allocation

We consider a team of autonomous agents that navigate in an adversarial environment and aim to achieve a task by allocating their resources over a set of target locations. The adversaries in the environment observe the autonomous team's behavior to infer their objective and counter-allocate their own resources to the target locations. In this setting, we develop strategies for controlling the density of the autonomous team so that they can deceive the adversaries regarding their objective while achieving the desired final resource allocation. We first develop a prediction algorithm, based on the principle of maximum entropy, to express the team's behavior expected by the adversaries. Then, by measuring the deceptiveness via Kullback-Leibler divergence, we develop convex optimization-based planning algorithms that deceives adversaries by either exaggerating the behavior towards a decoy allocation strategy or creating ambiguity regarding the final allocation strategy. Finally, we illustrate the performance of the proposed algorithms through numerical simulations.

Additive Logistic Mechanism for Privacy-Preserving Self-Supervised Learning

We study the privacy risks that are associated with training a neural network's weights with self-supervised learning algorithms. Through empirical evidence, we show that the fine-tuning stage, in which the network weights are updated with an informative and often private dataset, is vulnerable to privacy attacks. To address the vulnerabilities, we design a post-training privacy-protection algorithm that adds noise to the fine-tuned weights and propose a novel differential privacy mechanism that samples noise from the logistic distribution. Compared to the two conventional additive noise mechanisms, namely the Laplace and the Gaussian mechanisms, the proposed mechanism uses a bell-shaped distribution that resembles the distribution of the Gaussian mechanism, and it satisfies pure $\epsilon$-differential privacy similar to the Laplace mechanism. We apply membership inference attacks on both unprotected and protected models to quantify the trade-off between the models' privacy and performance. We show that the proposed protection algorithm can effectively reduce the attack accuracy to roughly 50\%-equivalent to random guessing-while maintaining a performance loss below 5\%.

Joint Learning of Reward Machines and Policies in Environments with Partially Known Semantics

We study the problem of reinforcement learning for a task encoded by a reward machine. The task is defined over a set of properties in the environment, called atomic propositions, and represented by Boolean variables. One unrealistic assumption commonly used in the literature is that the truth values of these propositions are accurately known. In real situations, however, these truth values are uncertain since they come from sensors that suffer from imperfections. At the same time, reward machines can be difficult to model explicitly, especially when they encode complicated tasks. We develop a reinforcement-learning algorithm that infers a reward machine that encodes the underlying task while learning how to execute it, despite the uncertainties of the propositions' truth values. In order to address such uncertainties, the algorithm maintains a probabilistic estimate about the truth value of the atomic propositions; it updates this estimate according to new sensory measurements that arrive from the exploration of the environment. Additionally, the algorithm maintains a hypothesis reward machine, which acts as an estimate of the reward machine that encodes the task to be learned. As the agent explores the environment, the algorithm updates the hypothesis reward machine according to the obtained rewards and the estimate of the atomic propositions' truth value. Finally, the algorithm uses a Q-learning procedure for the states of the hypothesis reward machine to determine the policy that accomplishes the task. We prove that the algorithm successfully infers the reward machine and asymptotically learns a policy that accomplishes the respective task.

Safe Reinforcement Learning via Shielding for POMDPs

Reinforcement learning (RL) in safety-critical environments requires an agent to avoid decisions with catastrophic consequences. Various approaches addressing the safety of RL exist to mitigate this problem. In particular, so-called shields provide formal safety guarantees on the behavior of RL agents based on (partial) models of the agents' environment. Yet, the state-of-the-art generally assumes perfect sensing capabilities of the agents, which is unrealistic in real-life applications. The standard models to capture scenarios with limited sensing are partially observable Markov decision processes (POMDPs). Safe RL for these models remains an open problem so far. We propose and thoroughly evaluate a tight integration of formally-verified shields for POMDPs with state-of-the-art deep RL algorithms and create an efficacious method that safely learns policies under partial observability. We empirically demonstrate that an RL agent using a shield, beyond being safe, converges to higher values of expected reward. Moreover, shielded agents need an order of magnitude fewer training episodes than unshielded agents, especially in challenging sparse-reward settings.

Dynamic Certification for Autonomous Systems

Autonomous systems are often deployed in complex sociotechnical environments, such as public roads, where they must behave safely and securely. Unlike many traditionally engineered systems, autonomous systems are expected to behave predictably in varying "open world" environmental contexts that cannot be fully specified formally. As a result, assurance about autonomous systems requires us to develop new certification methods and mathematical tools that can bound the uncertainty engendered by these diverse deployment scenarios, rather than relying on static tools.

Safely: Safe Stochastic Motion Planning Under Constrained Sensing via Duality

Consider a robot operating in an uncertain environment with stochastic, dynamic obstacles. Despite the clear benefits for trajectory optimization, it is often hard to keep track of each obstacle at every time step due to sensing and hardware limitations. We introduce the Safely motion planner, a receding-horizon control framework, that simultaneously synthesizes both a trajectory for the robot to follow as well as a sensor selection strategy that prescribes trajectory-relevant obstacles to measure at each time step while respecting the sensing constraints of the robot. We perform the motion planning using sequential quadratic programming, and prescribe obstacles to sense based on the duality information associated with the convex subproblems. We guarantee safety by ensuring that the probability of the robot colliding with any of the obstacles is below a prescribed threshold at every time step of the planned robot trajectory. We demonstrate the efficacy of the Safely motion planner through software and hardware experiments.

Hierarchical Control for Multi-Agent Autonomous Racing

We develop a hierarchical controller for multi-agent autonomous racing. A high-level planner approximates the race as a discrete game with simplified dynamics that encodes the complex safety and fairness rules seen in real-life racing and calculates a series of target waypoints. The low-level controller takes the resulting waypoints as a reference trajectory and computes high-resolution control inputs by solving a simplified formulation of a multi-agent racing game. We consider two approaches for the low-level planner to construct two hierarchical controllers. One approach uses multi-agent reinforcement learning (MARL), and the other solves a linear-quadratic Nash game (LQNG) to produce control inputs. We test the controllers against three baselines: an end-to-end MARL controller, a MARL controller tracking a fixed racing line, and an LQNG controller tracking a fixed racing line. Quantitative results show that the proposed hierarchical methods outperform their respective baseline methods in terms of head-to-head race wins and abiding by the rules. The hierarchical controller using MARL for low-level control consistently outperformed all other methods by winning over 88% of head-to-head races and more consistently adhered to the complex racing rules. Qualitatively, we observe the proposed controllers mimicking actions performed by expert human drivers such as shielding/blocking, overtaking, and long-term planning for delayed advantages. We show that hierarchical planning for game-theoretic reasoning produces competitive behavior even when challenged with complex rules and constraints.