Abhishek Gupta

Yuqing Du, Olivia Watkins, Zihan Wang et al. · microsoft-research, mit

Reinforcement learning algorithms typically struggle in the absence of a dense, well-shaped reward function. Intrinsically motivated exploration methods address this limitation by rewarding agents for visiting novel states or transitions, but these methods offer limited benefits in large environments where most discovered novelty is irrelevant for downstream tasks. We describe a method that uses background knowledge from text corpora to shape exploration. This method, called ELLM (Exploring with LLMs) rewards an agent for achieving goals suggested by a language model prompted with a description of the agent's current state. By leveraging large-scale language model pretraining, ELLM guides agents toward human-meaningful and plausibly useful behaviors without requiring a human in the loop. We evaluate ELLM in the Crafter game environment and the Housekeep robotic simulator, showing that ELLM-trained agents have better coverage of common-sense behaviors during pretraining and usually match or improve performance on a range of downstream tasks. Code available at https://github.com/yuqingd/ellm.

Zoey Qiuyu Chen, Karl Van Wyk, Yu-Wei Chao et al. · nvidia

Teaching a multi-fingered dexterous robot to grasp objects in the real world has been a challenging problem due to its high dimensional state and action space. We propose a robot-learning system that can take a small number of human demonstrations and learn to grasp unseen object poses given partially occluded observations. Our system leverages a small motion capture dataset and generates a large dataset with diverse and successful trajectories for a multi-fingered robot gripper. By adding domain randomization, we show that our dataset provides robust grasping trajectories that can be transferred to a policy learner. We train a dexterous grasping policy that takes the point clouds of the object as input and predicts continuous actions to grasp objects from different initial robot states. We evaluate the effectiveness of our system on a 22-DoF floating Allegro Hand in simulation and a 23-DoF Allegro robot hand with a KUKA arm in real world. The policy learned from our dataset can generalize well on unseen object poses in both simulation and the real world

Abhishek Gupta, Aldo Pacchiano, Yuexiang Zhai et al.

Reinforcement learning provides an automated framework for learning behaviors from high-level reward specifications, but in practice the choice of reward function can be crucial for good results -- while in principle the reward only needs to specify what the task is, in reality practitioners often need to design more detailed rewards that provide the agent with some hints about how the task should be completed. The idea of this type of ``reward-shaping'' has been often discussed in the literature, and is often a critical part of practical applications, but there is relatively little formal characterization of how the choice of reward shaping can yield benefits in sample complexity. In this work, we build on the framework of novelty-based exploration to provide a simple scheme for incorporating shaped rewards into RL along with an analysis tool to show that particular choices of reward shaping provably improve sample efficiency. We characterize the class of problems where these gains are expected to be significant and show how this can be connected to practical algorithms in the literature. We confirm that these results hold in practice in an experimental evaluation, providing an insight into the mechanisms through which reward shaping can significantly improve the complexity of reinforcement learning while retaining asymptotic performance.

Kelvin Xu, Zheyuan Hu, Ria Doshi et al.

Complex and contact-rich robotic manipulation tasks, particularly those that involve multi-fingered hands and underactuated object manipulation, present a significant challenge to any control method. Methods based on reinforcement learning offer an appealing choice for such settings, as they can enable robots to learn to delicately balance contact forces and dexterously reposition objects without strong modeling assumptions. However, running reinforcement learning on real-world dexterous manipulation systems often requires significant manual engineering. This negates the benefits of autonomous data collection and ease of use that reinforcement learning should in principle provide. In this paper, we describe a system for vision-based dexterous manipulation that provides a "programming-free" approach for users to define new tasks and enable robots with complex multi-fingered hands to learn to perform them through interaction. The core principle underlying our system is that, in a vision-based setting, users should be able to provide high-level intermediate supervision that circumvents challenges in teleoperation or kinesthetic teaching which allow a robot to not only learn a task efficiently but also to autonomously practice. Our system includes a framework for users to define a final task and intermediate sub-tasks with image examples, a reinforcement learning procedure that learns the task autonomously without interventions, and experimental results with a four-finger robotic hand learning multi-stage object manipulation tasks directly in the real world, without simulation, manual modeling, or reward engineering.

Max Simchowitz, Abhishek Gupta, Kaiqing Zhang

Obtaining rigorous statistical guarantees for generalization under distribution shift remains an open and active research area. We study a setting we call combinatorial distribution shift, where (a) under the test- and training-distributions, the labels $z$ are determined by pairs of features $(x,y)$, (b) the training distribution has coverage of certain marginal distributions over $x$ and $y$ separately, but (c) the test distribution involves examples from a product distribution over $(x,y)$ that is {not} covered by the training distribution. Focusing on the special case where the labels are given by bilinear embeddings into a Hilbert space $H$: $\mathbb{E}[z \mid x,y ]=\langle f_{\star}(x),g_{\star}(y)\rangle_{H}$, we aim to extrapolate to a test distribution domain that is $not$ covered in training, i.e., achieving bilinear combinatorial extrapolation. Our setting generalizes a special case of matrix completion from missing-not-at-random data, for which all existing results require the ground-truth matrices to be either exactly low-rank, or to exhibit very sharp spectral cutoffs. In this work, we develop a series of theoretical results that enable bilinear combinatorial extrapolation under gradual spectral decay as observed in typical high-dimensional data, including novel algorithms, generalization guarantees, and linear-algebraic results. A key tool is a novel perturbation bound for the rank-$k$ singular value decomposition approximations between two matrices that depends on the relative spectral gap rather than the absolute spectral gap, a result that may be of broader independent interest.

Balsam Dakhil, Abhishek Gupta

We analyze the problem of a renewable generator who wants to sell its random generation in a two-stage market to a number of strategic, but flexible, load serving entities (LSEs). To offer an incentive to participate in the auction, the generator promises to pay a penalty associated with any shortfall in generation due to the uncertain nature of the traded resource. We devise an auction mechanism that efficiently allocates electricity among LSEs while eliciting their true valuations. We show that the consumer surplus of the LSEs and expected profit of the generator are positive, thereby creating a win-win situation for all market participants.

Marcel Torne, Anthony Simeonov, Zechu Li et al.

Imitation learning methods need significant human supervision to learn policies robust to changes in object poses, physical disturbances, and visual distractors. Reinforcement learning, on the other hand, can explore the environment autonomously to learn robust behaviors but may require impractical amounts of unsafe real-world data collection. To learn performant, robust policies without the burden of unsafe real-world data collection or extensive human supervision, we propose RialTo, a system for robustifying real-world imitation learning policies via reinforcement learning in "digital twin" simulation environments constructed on the fly from small amounts of real-world data. To enable this real-to-sim-to-real pipeline, RialTo proposes an easy-to-use interface for quickly scanning and constructing digital twins of real-world environments. We also introduce a novel "inverse distillation" procedure for bringing real-world demonstrations into simulated environments for efficient fine-tuning, with minimal human intervention and engineering required. We evaluate RialTo across a variety of robotic manipulation problems in the real world, such as robustly stacking dishes on a rack, placing books on a shelf, and six other tasks. RialTo increases (over 67%) in policy robustness without requiring extensive human data collection. Project website and videos at https://real-to-sim-to-real.github.io/RialTo/

Sagnik Bhattacharya, Abhishek K. Gupta

An efficient channel estimation is of vital importance to help THz communication systems achieve their full potential. Conventional uplink channel estimation methods, such as least square estimation, are practically inefficient for THz systems because of their large computation overhead. In this paper, we propose an efficient convolutional neural network (CNN) based THz channel estimator that estimates the THz channel factors using uplink sub-6GHz channel. Further, we use the estimated THz channel factors to predict the optimal beamformer from a pre-given codebook, using a dense neural network. We not only get rid of the overhead associated with the conventional methods, but also achieve near-optimal spectral efficiency rates using the proposed beamformer predictor. The proposed method also outperforms deep learning based beamformer predictors accepting THz channel matrices as input, thus proving the validity and efficiency of our sub-6GHz based approach.

Lu Bai, Abhishek Gupta, Yew-Soon Ong

Multi-task learning solves multiple correlated tasks. However, conflicts may exist between them. In such circumstances, a single solution can rarely optimize all the tasks, leading to performance trade-offs. To arrive at a set of optimized yet well-distributed models that collectively embody different trade-offs in one algorithmic pass, this paper proposes to view Pareto multi-task learning through the lens of multi-task optimization. Multi-task learning is first cast as a multi-objective optimization problem, which is then decomposed into a diverse set of unconstrained scalar-valued subproblems. These subproblems are solved jointly using a novel multi-task gradient descent method, whose uniqueness lies in the iterative transfer of model parameters among the subproblems during the course of optimization. A theorem proving faster convergence through the inclusion of such transfers is presented. We investigate the proposed multi-task learning with multi-task optimization for solving various problem settings including image classification, scene understanding, and multi-target regression. Comprehensive experiments confirm that the proposed method significantly advances the state-of-the-art in discovering sets of Pareto-optimized models. Notably, on the large image dataset we tested on, namely NYUv2, the hypervolume convergence achieved by our method was found to be nearly two times faster than the next-best among the state-of-the-art.

Yuntian Deng, Xingyu Zhou, Baekjin Kim et al.

In this paper, we consider the Gaussian process (GP) bandit optimization problem in a non-stationary environment. To capture external changes, the black-box function is allowed to be time-varying within a reproducing kernel Hilbert space (RKHS). To this end, we develop WGP-UCB, a novel UCB-type algorithm based on weighted Gaussian process regression. A key challenge is how to cope with infinite-dimensional feature maps. To that end, we leverage kernel approximation techniques to prove a sublinear regret bound, which is the first (frequentist) sublinear regret guarantee on weighted time-varying bandits with general nonlinear rewards. This result generalizes both non-stationary linear bandits and standard GP-UCB algorithms. Further, a novel concentration inequality is achieved for weighted Gaussian process regression with general weights. We also provide universal upper bounds and weight-dependent upper bounds for weighted maximum information gains. These results are of independent interest for applications such as news ranking and adaptive pricing, where weights can be adopted to capture the importance or quality of data. Finally, we conduct experiments to highlight the favorable gains of the proposed algorithm in many cases when compared to existing methods.

Zhaoxuan Zhu, Nicola Pivaro, Shobhit Gupta et al.

Connected and Automated Hybrid Electric Vehicles have the potential to reduce fuel consumption and travel time in real-world driving conditions. The eco-driving problem seeks to design optimal speed and power usage profiles based upon look-ahead information from connectivity and advanced mapping features. Recently, Deep Reinforcement Learning (DRL) has been applied to the eco-driving problem. While the previous studies synthesize simulators and model-free DRL to reduce online computation, this work proposes a Safe Off-policy Model-Based Reinforcement Learning algorithm for the eco-driving problem. The advantages over the existing literature are three-fold. First, the combination of off-policy learning and the use of a physics-based model improves the sample efficiency. Second, the training does not require any extrinsic rewarding mechanism for constraint satisfaction. Third, the feasibility of trajectory is guaranteed by using a safe set approximated by deep generative models. The performance of the proposed method is benchmarked against a baseline controller representing human drivers, a previously designed model-free DRL strategy, and the wait-and-see optimal solution. In simulation, the proposed algorithm leads to a policy with a higher average speed and a better fuel economy compared to the model-free agent. Compared to the baseline controller, the learned strategy reduces the fuel consumption by more than 21\% while keeping the average speed comparable.

Abhishek Gupta, William B. Haskell

This paper develops a unified framework, based on iterated random operator theory, to analyze the convergence of constant stepsize recursive stochastic algorithms (RSAs). RSAs use randomization to efficiently compute expectations, and so their iterates form a stochastic process. The key idea of our analysis is to lift the RSA into an appropriate higher-dimensional space and then express it as an equivalent Markov chain. Instead of determining the convergence of this Markov chain (which may not converge under constant stepsize), we study the convergence of the distribution of this Markov chain. To study this, we define a new notion of Wasserstein divergence. We show that if the distribution of the iterates in the Markov chain satisfy a contraction property with respect to the Wasserstein divergence, then the Markov chain admits an invariant distribution. We show that convergence of a large family of constant stepsize RSAs can be understood using this framework, and we provide several detailed examples.

Anant Chopra, Deepak S. Kalhan, Amrit S. Bedi et al.

Traffic assignment is an integral part of urban city planning. Roads and freeways are constructed to cater to the expected demands of the commuters between different origin-destination pairs with the overall objective of minimising the travel cost. As compared to static traffic assignment problems where the traffic network is fixed over time, a dynamic traffic network is more realistic where the network's cost parameters change over time due to the presence of random congestion. In this paper, we consider a stochastic version of the traffic assignment problem where the central planner is interested in finding an optimal social flow in the presence of random users. These users are random and cannot be controlled by any central directives. We propose a Frank-Wolfe algorithm based stochastic algorithm to determine the socially optimal flow for the stochastic setting in an online manner. Further, simulation results corroborate the efficacy of the proposed algorithm.

Abhishek Gupta, Rahul Jain, Peter Glynn

In many branches of engineering, Banach contraction mapping theorem is employed to establish the convergence of certain deterministic algorithms. Randomized versions of these algorithms have been developed that have proved useful in data-driven problems. In a class of randomized algorithms, in each iteration, the contraction map is approximated with an operator that uses independent and identically distributed samples of certain random variables. This leads to iterated random operators acting on an initial point in a complete metric space, and it generates a Markov chain. In this paper, we develop a new stochastic dominance based proof technique, called probabilistic contraction analysis, for establishing the convergence in probability of Markov chains generated by such iterated random operators in certain limiting regime. The methods developed in this paper provides a general framework for understanding convergence of a wide variety of Monte Carlo methods in which contractive property is present. We apply the convergence result to conclude the convergence of fitted value iteration and fitted relative value iteration in continuous state and continuous action Markov decision problems as representative applications of the general framework developed here.

Aravind Rajeswaran, Vikash Kumar, Abhishek Gupta et al.

Dexterous multi-fingered hands are extremely versatile and provide a generic way to perform a multitude of tasks in human-centric environments. However, effectively controlling them remains challenging due to their high dimensionality and large number of potential contacts. Deep reinforcement learning (DRL) provides a model-agnostic approach to control complex dynamical systems, but has not been shown to scale to high-dimensional dexterous manipulation. Furthermore, deployment of DRL on physical systems remains challenging due to sample inefficiency. Consequently, the success of DRL in robotics has thus far been limited to simpler manipulators and tasks. In this work, we show that model-free DRL can effectively scale up to complex manipulation tasks with a high-dimensional 24-DoF hand, and solve them from scratch in simulated experiments. Furthermore, with the use of a small number of human demonstrations, the sample complexity can be significantly reduced, which enables learning with sample sizes equivalent to a few hours of robot experience. The use of demonstrations result in policies that exhibit very natural movements and, surprisingly, are also substantially more robust.