We study Policy-extended Value Function Approximator (PeVFA) in Reinforcement Learning (RL), which extends conventional value function approximator (VFA) to take as input not only the state (and action) but also an explicit policy representation. Such an extension enables PeVFA to preserve values of multiple policies at the same time and brings an appealing characteristic, i.e., value generalization among policies. We formally analyze the value generalization under Generalized Policy Iteration (GPI). From theoretical and empirical lens, we show that generalized value estimates offered by PeVFA may have lower initial approximation error to true values of successive policies, which is expected to improve consecutive value approximation during GPI. Based on above clues, we introduce a new form of GPI with PeVFA which leverages the value generalization along policy improvement path. Moreover, we propose a representation learning framework for RL policy, providing several approaches to learn effective policy embeddings from policy network parameters or state-action pairs. In our experiments, we evaluate the efficacy of value generalization offered by PeVFA and policy representation learning in several OpenAI Gym continuous control tasks. For a representative instance of algorithm implementation, Proximal Policy Optimization (PPO) re-implemented under the paradigm of GPI with PeVFA achieves about 40% performance improvement on its vanilla counterpart in most environments.

We consider real-world reinforcement learning (RL) of robotic manipulation tasks that involve both visuomotor skills and contact-rich skills. We aim to train a policy that maps multimodal sensory observations (vision and force) to a manipulator's joint velocities under practical considerations. We propose to use offline samples to learn a set of general value functions (GVFs) that make counterfactual predictions from the visual inputs. We show that combining the offline learned counterfactual predictions with force feedbacks in online policy learning allows efficient reinforcement learning given only a terminal (success/failure) reward. We argue that the learned counterfactual predictions form a compact and informative representation that enables sample efficiency and provides auxiliary reward signals that guide online explorations towards contact-rich states. Various experiments in simulation and real-world settings were performed for evaluation. Recordings of the real-world robot training can be found via https://sites.google.com/view/realrl.

We propose a multi-time-scale predictive representation learning method to efficiently learn robust driving policies in an offline manner that generalize well to novel road geometries, and damaged and distracting lane conditions which are not covered in the offline training data. We show that our proposed representation learning method can be applied easily in an offline (batch) reinforcement learning setting demonstrating the ability to generalize well and efficiently under novel conditions compared to standard batch RL methods. Our proposed method utilizes training data collected entirely offline in the real-world which removes the need of intensive online explorations that impede applying deep reinforcement learning on real-world robot training. Various experiments were conducted in both simulator and real-world scenarios for the purpose of evaluation and analysis of our proposed claims.

We explore fixed-horizon temporal difference (TD) methods, reinforcement learning algorithms for a new kind of value function that predicts the sum of rewards over a fixed number of future time steps. To learn the value function for horizon h, these algorithms bootstrap from the value function for horizon h−1, or some shorter horizon. Because no value function bootstraps from itself, fixed-horizon methods are immune to the stability problems that plague other off-policy TD methods using function approximation (also known as “the deadly triad”). Although fixed-horizon methods require the storage of additional value functions, this gives the agent additional predictive power, while the added complexity can be substantially reduced via parallel updates, shared weights, and n-step bootstrapping. We show how to use fixed-horizon value functions to solve reinforcement learning problems competitively with methods such as Q-learning that learn conventional value functions. We also prove convergence of fixed-horizon temporal difference methods with linear and general function approximation. Taken together, our results establish fixed-horizon TD methods as a viable new way of avoiding the stability problems of the deadly triad.

We consider the problem of mapless collision-avoidance navigation where humans are present using 2D laser scans. Our proposed method uses ego-safety to measure collision from the robot's perspective and social-safety to measure the impact of robot's actions on surrounding pedestrians. Specifically, the social-safety part predicts the intrusion impact of the robot's action into the interaction area with surrounding humans. We train the policy using reinforcement learning on a simple simulator and directly evaluate the learned policy in Gazebo and real robot tests. Experiments show the learned policy smoothly transferred to different scenarios without any fine tuning. We observe that our method demonstrates time-efficient path planning behavior with high success rate in the mapless navigation task. Furthermore, we test our method in a navigation task among dynamic crowds, considering both low and high volume traffic. Our learned policy demonstrates cooperative behavior that actively drives our robot into traffic flows while showing respect to nearby pedestrians. Evaluation videos are at https://sites.google.com/view/ssw-batman.

We propose and demonstrate a framework called perception as prediction for autonomous driving that uses general value functions (GVFs) to learn predictions. Perception as prediction learns data-driven predictions relating to the impact of actions on the agent's perception of the world. It also provides a data-driven approach to predict the impact of the anticipated behavior of other agents on the world without explicitly learning their policy or intentions. We demonstrate perception as prediction by learning to predict an agent's front safety and rear safety with GVFs, which encapsulate anticipation of the behavior of the vehicle in front and in the rear, respectively. The safety predictions are learned through random interactions in a simulated environment containing other agents. We show that these predictions can be used to produce similar control behavior to an LQR-based controller in an adaptive cruise control problem as well as provide advanced warning when the vehicle behind is approaching dangerously. The predictions are compact policy-based predictions that support prediction of the long term impact on safety when following a given policy. We analyze two controllers that use the learned predictions in a racing simulator to understand the value of the predictions and demonstrate their use in the real-world on a Clearpath Jackal robot and an autonomous vehicle platform.

Importance sampling (IS) is a common reweighting strategy for off-policy prediction in reinforcement learning. While it is consistent and unbiased, it can result in high variance updates to the weights for the value function. In this work, we explore a resampling strategy as an alternative to reweighting. We propose Importance Resampling (IR) for off-policy prediction, which resamples experience from a replay buffer and applies standard on-policy updates. The approach avoids using importance sampling ratios in the update, instead correcting the distribution before the update. We characterize the bias and consistency of IR, particularly compared to Weighted IS (WIS). We demonstrate in several microworlds that IR has improved sample efficiency and lower variance updates, as compared to IS and several variance-reduced IS strategies, including variants of WIS and V-trace which clips IS ratios. We also provide a demonstration showing IR improves over IS for learning a value function from images in a racing car simulator.

Summary This paper presents a scalable object detection workflow for detecting objects, such as settlements, from remotely sensed (RS) imagery. We have successfully deployed this workflow on Titan supercomputer and utilized it for the task of mapping human settlement at a country scale. The performance of various stages in the workflow was analyzed before making it operational. The workflow implemented various strategies to address issues such as suboptimal resource utilization and long‐tail effects due to unbalanced image workload, data loss due to runtime failures, and maximum wall‐time constraints imposed by Titan's job scheduling policy. A mean shift clustering–based static load balancing strategy was implemented, which partitions the image load such that each partition contained similar‐sized images. Furthermore, a checkpoint‐restart strategy was added in the workflow as a fault‐tolerance mechanism to prevent the data losses due to unforeseen runtime failures. The performance of the above‐mentioned strategies was observed in various scenarios, such as node failure, exceeding wall time, and successful completion. Using this workflow, we have processed an RS data set that has a spatial resolution of 0.31 m and is comprised of 685 675 km2 of area of the Republic of Zambia in under six hours using 5426 nodes of the Titan supercomputer.

In order to deal with the sequential decision problems with large or continuous state spaces, feature representation and function approximation have been a major research topic in reinforcement learning (RL). In this paper, a clustering-based graph Laplacian framework is presented for feature representation and value function approximation (VFA) in RL. By making use of clustering-based techniques, that is, K-means clustering or fuzzy C-means clustering, a graph Laplacian is constructed by subsampling in Markov decision processes (MDPs) with continuous state spaces. The basis functions for VFA can be automatically generated from spectral analysis of the graph Laplacian. The clustering-based graph Laplacian is integrated with a class of approximation policy iteration algorithms called representation policy iteration (RPI) for RL in MDPs with continuous state spaces. Simulation and experimental results show that, compared with previous RPI methods, the proposed approach needs fewer sample points to compute an efficient set of basis functions and the learning control performance can be improved for a variety of parameter settings.

A new time series prediction architecture is introduced using a fuzzy inference system (FIS) and a new framework for fuzzy relational clustering of time series. The FIS is used to predict future samples in a time series where recurrent neural networks comprise the consequents of the rules. The antecedents come in the form of fuzzy relations; however, previous approaches such as FCM build these antecedents in a Euclidean feature space which is very limiting and not well suited to the problem of clustering time series. Our approach to learning the antecedents of the rules involves clustering time series using proximity values, indicative of closeness. A variant of the classical correlation is used to measure proximity. Our objective is to investigate and evaluate the application of proximity fuzzy clustering in the domain of time series prediction by comparing its performance against several commonly used time series prediction models.