Rating-based Reinforcement Learning

This paper develops a novel rating-based reinforcement learning approach that uses human ratings to obtain human guidance in reinforcement learning. Different from the existing preference-based and ranking-based reinforcement learning paradigms, based on human relative preferences over sample pairs, the proposed rating-based reinforcement learning approach is based on human evaluation of individual trajectories without relative comparisons between sample pairs. The rating-based reinforcement learning approach builds on a new prediction model for human ratings and a novel multi-class loss function. We conduct several experimental studies based on synthetic ratings and real human ratings to evaluate the effectiveness and benefits of the new rating-based reinforcement learning approach.

DIP-RL: Demonstration-Inferred Preference Learning in Minecraft

In machine learning for sequential decision-making, an algorithmic agent learns to interact with an environment while receiving feedback in the form of a reward signal. However, in many unstructured real-world settings, such a reward signal is unknown and humans cannot reliably craft a reward signal that correctly captures desired behavior. To solve tasks in such unstructured and open-ended environments, we present Demonstration-Inferred Preference Reinforcement Learning (DIP-RL), an algorithm that leverages human demonstrations in three distinct ways, including training an autoencoder, seeding reinforcement learning (RL) training batches with demonstration data, and inferring preferences over behaviors to learn a reward function to guide RL. We evaluate DIP-RL in a tree-chopping task in Minecraft. Results suggest that the method can guide an RL agent to learn a reward function that reflects human preferences and that DIP-RL performs competitively relative to baselines. DIP-RL is inspired by our previous work on combining demonstrations and pairwise preferences in Minecraft, which was awarded a research prize at the 2022 NeurIPS MineRL BASALT competition, Learning from Human Feedback in Minecraft. Example trajectory rollouts of DIP-RL and baselines are located at https://sites.google.com/view/dip-rl.

Towards Solving Fuzzy Tasks with Human Feedback: A Retrospective of the MineRL BASALT 2022 Competition

To facilitate research in the direction of fine-tuning foundation models from human feedback, we held the MineRL BASALT Competition on Fine-Tuning from Human Feedback at NeurIPS 2022. The BASALT challenge asks teams to compete to develop algorithms to solve tasks with hard-to-specify reward functions in Minecraft. Through this competition, we aimed to promote the development of algorithms that use human feedback as channels to learn the desired behavior. We describe the competition and provide an overview of the top solutions. We conclude by discussing the impact of the competition and future directions for improvement.

Efficient Preference-Based Reinforcement Learning Using Learned Dynamics Models

Preference-based reinforcement learning (PbRL) can enable robots to learn to perform tasks based on an individual's preferences without requiring a hand-crafted reward function. However, existing approaches either assume access to a high-fidelity simulator or analytic model or take a model-free approach that requires extensive, possibly unsafe online environment interactions. In this paper, we study the benefits and challenges of using a learned dynamics model when performing PbRL. In particular, we provide evidence that a learned dynamics model offers the following benefits when performing PbRL: (1) preference elicitation and policy optimization require significantly fewer environment interactions than model-free PbRL, (2) diverse preference queries can be synthesized safely and efficiently as a byproduct of standard model-based RL, and (3) reward pre-training based on suboptimal demonstrations can be performed without any environmental interaction. Our paper provides empirical evidence that learned dynamics models enable robots to learn customized policies based on user preferences in ways that are safer and more sample efficient than prior preference learning approaches.

Autonomously Untangling Long Cables

Cables are ubiquitous in many settings, but are prone to self-occlusions and knots, making them difficult to perceive and manipulate. The challenge often increases with cable length: long cables require more complex slack management and strategies to facilitate observability and reachability. In this paper, we focus on autonomously untangling cables up to 3 meters in length using a bilateral robot. We develop new motion primitives to efficiently untangle long cables and novel gripper jaws specialized for this task. We present Sliding and Grasping for Tangle Manipulation (SGTM), an algorithm that composes these primitives with RGBD vision to iteratively untangle. SGTM untangles cables with success rates of 67% on isolated overhand and figure eight knots and 50% on more complex configurations. Supplementary material, visualizations, and videos can be found at https://sites.google.com/view/rss-2022-untangling/home.

Learning Switching Criteria for Sim2Real Transfer of Robotic Fabric Manipulation Policies

Simulation-to-reality transfer has emerged as a popular and highly successful method to train robotic control policies for a wide variety of tasks. However, it is often challenging to determine when policies trained in simulation are ready to be transferred to the physical world. Deploying policies that have been trained with very little simulation data can result in unreliable and dangerous behaviors on physical hardware. On the other hand, excessive training in simulation can cause policies to overfit to the visual appearance and dynamics of the simulator. In this work, we study strategies to automatically determine when policies trained in simulation can be reliably transferred to a physical robot. We specifically study these ideas in the context of robotic fabric manipulation, in which successful sim2real transfer is especially challenging due to the difficulties of precisely modeling the dynamics and visual appearance of fabric. Results in a fabric smoothing task suggest that our switching criteria correlate well with performance in real. In particular, our confidence-based switching criteria achieve average final fabric coverage of 87.2-93.7% within 55-60% of the total training budget. See https://tinyurl.com/lsc-case for code and supplemental materials.

Efficiently Learning Single-Arm Fling Motions to Smooth Garments

Recent work has shown that 2-arm "fling" motions can be effective for garment smoothing. We consider single-arm fling motions. Unlike 2-arm fling motions, which require little robot trajectory parameter tuning, single-arm fling motions are sensitive to trajectory parameters. We consider a single 6-DOF robot arm that learns fling trajectories to achieve high garment coverage. Given a garment grasp point, the robot explores different parameterized fling trajectories in physical experiments. To improve learning efficiency, we propose a coarse-to-fine learning method that first uses a multi-armed bandit (MAB) framework to efficiently find a candidate fling action, which it then refines via a continuous optimization method. Further, we propose novel training and execution-time stopping criteria based on fling outcome uncertainty. Compared to baselines, we show that the proposed method significantly accelerates learning. Moreover, with prior experience on similar garments collected through self-supervision, the MAB learning time for a new garment is reduced by up to 87%. We evaluate on 6 garment types: towels, T-shirts, long-sleeve shirts, dresses, sweat pants, and jeans. Results suggest that using prior experience, a robot requires under 30 minutes to learn a fling action for a novel garment that achieves 60-94% coverage.

Policy-Based Bayesian Experimental Design for Non-Differentiable Implicit Models

For applications in healthcare, physics, energy, robotics, and many other fields, designing maximally informative experiments is valuable, particularly when experiments are expensive, time-consuming, or pose safety hazards. While existing approaches can sequentially design experiments based on prior observation history, many of these methods do not extend to implicit models, where simulation is possible but computing the likelihood is intractable. Furthermore, they often require either significant online computation during deployment or a differentiable simulation system. We introduce Reinforcement Learning for Deep Adaptive Design (RL-DAD), a method for simulation-based optimal experimental design for non-differentiable implicit models. RL-DAD extends prior work in policy-based Bayesian Optimal Experimental Design (BOED) by reformulating it as a Markov Decision Process with a reward function based on likelihood-free information lower bounds, which is used to learn a policy via deep reinforcement learning. The learned design policy maps prior histories to experiment designs offline and can be quickly deployed during online execution. We evaluate RL-DAD and find that it performs competitively with baselines on three benchmarks.

ThriftyDAgger: Budget-Aware Novelty and Risk Gating for Interactive Imitation Learning

Effective robot learning often requires online human feedback and interventions that can cost significant human time, giving rise to the central challenge in interactive imitation learning: is it possible to control the timing and length of interventions to both facilitate learning and limit burden on the human supervisor? This paper presents ThriftyDAgger, an algorithm for actively querying a human supervisor given a desired budget of human interventions. ThriftyDAgger uses a learned switching policy to solicit interventions only at states that are sufficiently (1) novel, where the robot policy has no reference behavior to imitate, or (2) risky, where the robot has low confidence in task completion. To detect the latter, we introduce a novel metric for estimating risk under the current robot policy. Experiments in simulation and on a physical cable routing experiment suggest that ThriftyDAgger's intervention criteria balances task performance and supervisor burden more effectively than prior algorithms. ThriftyDAgger can also be applied at execution time, where it achieves a 100% success rate on both the simulation and physical tasks. A user study (N=10) in which users control a three-robot fleet while also performing a concentration task suggests that ThriftyDAgger increases human and robot performance by 58% and 80% respectively compared to the next best algorithm while reducing supervisor burden.

Untangling Dense Non-Planar Knots by Learning Manipulation Features and Recovery Policies

Robot manipulation for untangling 1D deformable structures such as ropes, cables, and wires is challenging due to their infinite dimensional configuration space, complex dynamics, and tendency to self-occlude. Analytical controllers often fail in the presence of dense configurations, due to the difficulty of grasping between adjacent cable segments. We present two algorithms that enhance robust cable untangling, LOKI and SPiDERMan, which operate alongside HULK, a high-level planner from prior work. LOKI uses a learned model of manipulation features to refine a coarse grasp keypoint prediction to a precise, optimized location and orientation, while SPiDERMan uses a learned model to sense task progress and apply recovery actions. We evaluate these algorithms in physical cable untangling experiments with 336 knots and over 1500 actions on real cables using the da Vinci surgical robot. We find that the combination of HULK, LOKI, and SPiDERMan is able to untangle dense overhand, figure-eight, double-overhand, square, bowline, granny, stevedore, and triple-overhand knots. The composition of these methods successfully untangles a cable from a dense initial configuration in 68.3% of 60 physical experiments and achieves 50% higher success rates than baselines from prior work. Supplementary material, code, and videos can be found at https://tinyurl.com/rssuntangling.