We study the problem of learning controllers for discrete-time non-linear stochastic dynamical systems with formal reach-avoid guarantees. This work presents the first method for providing formal reach-avoid guarantees, which combine and generalize stability and safety guarantees, with a tolerable probability threshold $p\in[0,1]$ over the infinite time horizon. Our method leverages advances in machine learning literature and it represents formal certificates as neural networks. In particular, we learn a certificate in the form of a reach-avoid supermartingale (RASM), a novel notion that we introduce in this work. Our RASMs provide reachability and avoidance guarantees by imposing constraints on what can be viewed as a stochastic extension of level sets of Lyapunov functions for deterministic systems. Our approach solves several important problems -- it can be used to learn a control policy from scratch, to verify a reach-avoid specification for a fixed control policy, or to fine-tune a pre-trained policy if it does not satisfy the reach-avoid specification. We validate our approach on $3$ stochastic non-linear reinforcement learning tasks.
相關內容
We consider the linear contextual multi-class multi-period packing problem~(LMMP) where the goal is to pack items such that the total vector of consumption is below a given budget vector and the total value is as large as possible. We consider the setting where the reward and the consumption vector associated with each action is a class-dependent linear function of the context, and the decision-maker receives bandit feedback. LMMP includes linear contextual bandits with knapsacks and online revenue management as special cases. We establish a new more efficient estimator which guarantees a faster convergence rate, and consequently, a lower regret in such problems. We propose a bandit policy that is a closed-form function of said estimated parameters. When the contexts are non-degenerate, the regret of the proposed policy is sublinear in the context dimension, the number of classes, and the time horizon~$T$ when the budget grows at least as $\sqrt{T}$. We also resolve an open problem posed in Agrawal & Devanur (2016), and extend the result to a multi-class setting. Our numerical experiments clearly demonstrate that the performance of our policy is superior to other benchmarks in the literature.
Robustness and safety are critical for the trustworthy deployment of deep reinforcement learning in real-world decision making applications. In particular, we require algorithms that can guarantee robust, safe performance in the presence of general environment disturbances, while making limited assumptions on the data collection process during training. In this work, we propose a safe reinforcement learning framework with robustness guarantees through the use of an optimal transport cost uncertainty set. We provide an efficient, theoretically supported implementation based on Optimal Transport Perturbations, which can be applied in a completely offline fashion using only data collected in a nominal training environment. We demonstrate the robust, safe performance of our approach on a variety of continuous control tasks with safety constraints in the Real-World Reinforcement Learning Suite.
We investigate the problem of stochastic, combinatorial multi-armed bandits where the learner only has access to bandit feedback and the reward function can be non-linear. We provide a general framework for adapting discrete offline approximation algorithms into sublinear $\alpha$-regret methods that only require bandit feedback, achieving $\mathcal{O}\left(T^\frac{2}{3}\log(T)^\frac{1}{3}\right)$ expected cumulative $\alpha$-regret dependence on the horizon $T$. The framework only requires the offline algorithms to be robust to small errors in function evaluation. The adaptation procedure does not even require explicit knowledge of the offline approximation algorithm -- the offline algorithm can be used as black box subroutine. To demonstrate the utility of the proposed framework, the proposed framework is applied to multiple problems in submodular maximization, adapting approximation algorithms for cardinality and for knapsack constraints. The new CMAB algorithms for knapsack constraints outperform a full-bandit method developed for the adversarial setting in experiments with real-world data.
Neural Networks (NNs) have been successfully employed to represent the state evolution of complex dynamical systems. Such models, referred to as NN dynamic models (NNDMs), use iterative noisy predictions of NN to estimate a distribution of system trajectories over time. Despite their accuracy, safety analysis of NNDMs is known to be a challenging problem and remains largely unexplored. To address this issue, in this paper, we introduce a method of providing safety guarantees for NNDMs. Our approach is based on stochastic barrier functions, whose relation with safety are analogous to that of Lyapunov functions with stability. We first show a method of synthesizing stochastic barrier functions for NNDMs via a convex optimization problem, which in turn provides a lower bound on the system's safety probability. A key step in our method is the employment of the recent convex approximation results for NNs to find piece-wise linear bounds, which allow the formulation of the barrier function synthesis problem as a sum-of-squares optimization program. If the obtained safety probability is above the desired threshold, the system is certified. Otherwise, we introduce a method of generating controls for the system that robustly maximizes the safety probability in a minimally-invasive manner. We exploit the convexity property of the barrier function to formulate the optimal control synthesis problem as a linear program. Experimental results illustrate the efficacy of the method. Namely, they show that the method can scale to multi-dimensional NNDMs with multiple layers and hundreds of neurons per layer, and that the controller can significantly improve the safety probability.
We study reinforcement learning with linear function approximation and adversarially changing cost functions, a setup that has mostly been considered under simplifying assumptions such as full information feedback or exploratory conditions.We present a computationally efficient policy optimization algorithm for the challenging general setting of unknown dynamics and bandit feedback, featuring a combination of mirror-descent and least squares policy evaluation in an auxiliary MDP used to compute exploration bonuses.Our algorithm obtains an $\widetilde O(K^{6/7})$ regret bound, improving significantly over previous state-of-the-art of $\widetilde O (K^{14/15})$ in this setting. In addition, we present a version of the same algorithm under the assumption a simulator of the environment is available to the learner (but otherwise no exploratory assumptions are made), and prove it obtains state-of-the-art regret of $\widetilde O (K^{2/3})$.
Many instances of similar or almost-identical industrial machines or tools are often deployed at once, or in quick succession. For instance, a particular model of air compressor may be installed at hundreds of customers. Because these tools perform distinct but highly similar tasks, it is interesting to be able to quickly produce a high-quality controller for machine $N+1$ given the controllers already produced for machines $1..N$. This is even more important when the controllers are learned through Reinforcement Learning, as training takes time, energy and other resources. In this paper, we apply Policy Intersection, a Policy Shaping method, to help a Reinforcement Learning agent learn to solve a new variant of a compressors control problem faster, by transferring knowledge from several previously learned controllers. We show that our approach outperforms loading an old controller, and significantly improves performance in the long run.
In this paper, we study fast first-order algorithms that approximately solve linear programs (LPs). More specifically, we apply algorithms from online linear programming to offline LPs and derive algorithms that are free of any matrix multiplication. To further improve the applicability of the proposed methods, we propose a variable-duplication technique that achieves $\mathcal{O}(\sqrt{mn/K})$ optimality gap by copying each variable $K$ times. Moreover, we identify that online algorithms can be efficiently incorporated into a column generation framework for large-scale LPs. Finally, numerical experiments show that our proposed methods can be applied either as an approximate direct solver or as an initialization subroutine in frameworks of exact LP solving.
This note addresses the problem of evaluating the impact of an attack on discrete-time nonlinear stochastic control systems. The problem is formulated as an optimal control problem with a joint chance constraint that forces the adversary to avoid detection throughout a given time period. Due to the joint constraint, the optimal control policy depends not only on the current state, but also on the entire history, leading to an explosion of the search space and making the problem generally intractable. However, we discover that the current state and whether an alarm has been triggered, or not, is sufficient for specifying the optimal decision at each time step. This information, which we refer to as the alarm flag, can be added to the state space to create an equivalent optimal control problem that can be solved with existing numerical approaches using a Markov policy. Additionally, we note that the formulation results in a policy that does not avoid detection once an alarm has been triggered. We extend the formulation to handle multi-alarm avoidance policies for more reasonable attack impact evaluations, and show that the idea of augmenting the state space with an alarm flag is valid in this extended formulation as well.
Stochastic versions of proximal methods have gained much attention in statistics and machine learning. These algorithms tend to admit simple, scalable forms, and enjoy numerical stability via implicit updates. In this work, we propose and analyze a stochastic version of the recently proposed proximal distance algorithm, a class of iterative optimization methods that recover a desired constrained estimation problem as a penalty parameter $\rho \rightarrow \infty$. By uncovering connections to related stochastic proximal methods and interpreting the penalty parameter as the learning rate, we justify heuristics used in practical manifestations of the proximal distance method, establishing their convergence guarantees for the first time. Moreover, we extend recent theoretical devices to establish finite error bounds and a complete characterization of convergence rates regimes. We validate our analysis via a thorough empirical study, also showing that unsurprisingly, the proposed method outpaces batch versions on popular learning tasks.
Sampling methods (e.g., node-wise, layer-wise, or subgraph) has become an indispensable strategy to speed up training large-scale Graph Neural Networks (GNNs). However, existing sampling methods are mostly based on the graph structural information and ignore the dynamicity of optimization, which leads to high variance in estimating the stochastic gradients. The high variance issue can be very pronounced in extremely large graphs, where it results in slow convergence and poor generalization. In this paper, we theoretically analyze the variance of sampling methods and show that, due to the composite structure of empirical risk, the variance of any sampling method can be decomposed into \textit{embedding approximation variance} in the forward stage and \textit{stochastic gradient variance} in the backward stage that necessities mitigating both types of variance to obtain faster convergence rate. We propose a decoupled variance reduction strategy that employs (approximate) gradient information to adaptively sample nodes with minimal variance, and explicitly reduces the variance introduced by embedding approximation. We show theoretically and empirically that the proposed method, even with smaller mini-batch sizes, enjoys a faster convergence rate and entails a better generalization compared to the existing methods.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191