Automatic tuning of hyper-parameters of reinforcement learning algorithms using Bayesian optimization with behavioral cloning
Universidad Tecnológica Nacional（国立科技大学）；GISIQ, Facultad Regional Villa María, UTN（国立科技大学维拉玛丽亚地区学院 GISIQ）；CIT Villa María, CONICET-UNVM（国家科学与技术研究理事会-国立维拉玛丽亚大学维拉玛丽亚科技研究中心）；INGAR, CONICET-UTN（国家科学与技术研究理事会-国立科技大学工业自动化研究所）
Juan Cruz Barsce, Jorge A. Palombarini, Ernesto C. Martínez
Optimal setting of several hyper-parameters in machine learning algorithms is key to make the most of available data. To this aim, several methods such as evolutionary strategies, random search, Bayesian optimization and heuristic rules of thumb have been proposed. In reinforcement learning (RL), the information content of data gathered by the learning agent while interacting with its environment is heavily dependent on the setting of many hyper-parameters. Therefore, the user of an RL algorithm has to rely on search-based optimization methods, such as grid search or the Nelder-Mead simplex algorithm, that are very inefficient for most RL tasks, slows down significantly the learning curve and leaves to the user the burden of purposefully biasing data gathering. In this work, in order to make an RL algorithm more user-independent, a novel approach for autonomous hyper-parameter setting using Bayesian optimization is proposed. Data from past episodes and different hyper-parameter values are used at a meta-learning level by performing behavioral cloning which helps improving the effectiveness in maximizing a reinforcement learning variant of an acquisition function. Also, by tightly integrating Bayesian optimization in a reinforcement learning agent design, the number of state transitions needed to converge to the optimal policy for a given task is reduced. Computational experiments reveal promising results compared to other manual tweaking and optimization-based approaches which highlights the benefits of changing the algorithm hyper-parameters to increase the information content of generated data.
We introduce an offline reinforcement learning (RL) algorithm that explicitly clones a behavior policy to constrain value learning. In offline RL, it is often important to prevent a policy from selecting unobserved actions, since the consequence of these actions cannot be presumed without additional information about the environment. One straightforward way to implement such a constraint is to explicitly model a given data distribution via behavior cloning and directly force a policy not to select uncertain actions. However, many offline RL methods instantiate the constraint indirectly -- for …
Offline reinforcement learning (RL) algorithms can acquire effective policies by utilizing previously collected experience, without any online interaction. It is widely understood that offline RL is able to extract good policies even from highly suboptimal data, a scenario where imitation learning finds suboptimal solutions that do not improve over the demonstrator that generated the dataset. However, another common use case for practitioners is to learn from data that resembles demonstrations. In this case, one can choose to apply offline RL, but can also use behavioral cloning (BC) algorithm…
Behavioral cloning (BC) provides a straightforward solution to offline RL by mimicking offline trajectories via supervised learning. Recent advances (Chen et al., 2021; Janner et al., 2021; Emmons et al., 2021) have shown that by conditioning on desired future returns, BC can perform competitively to their value-based counterparts, while enjoying much more simplicity and training stability. While promising, we show that these methods can be unreliable, as their performance may degrade significantly when conditioned on high, out-of-distribution (ood) returns. This is crucial in practice, as we …
We find that across a wide range of robot policy learning scenarios, treating supervised policy learning with an implicit model generally performs better, on average, than commonly used explicit models. We present extensive experiments on this finding, and we provide both intuitive insight and theoretical arguments distinguishing the properties of implicit models compared to their explicit counterparts, particularly with respect to approximating complex, potentially discontinuous and multi-valued (set-valued) functions. On robotic policy learning tasks we show that implicit behavioral cloning …
Nanjing University（南京大学）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Polixir Technologies（波利克斯尔科技）
Imitation learning trains a policy by mimicking expert demonstrations. Various imitation methods were proposed and empirically evaluated, meanwhile, their theoretical understanding needs further studies. In this paper, we firstly analyze the value gap between the expert policy and imitated policies by two imitation methods, behavioral cloning and generative adversarial imitation. The results support that generative adversarial imitation can reduce the compounding errors compared to behavioral cloning, and thus has a better sample complexity. Noticed that by considering the environment transiti…
Imitation learning trains a policy from expert demonstrations. Imitation learning approaches have been designed from various principles, such as behavioral cloning via supervised learning, apprenticeship learning via inverse reinforcement learning, and GAIL via generative adversarial learning. In this paper, we propose a framework to analyze the theoretical property of imitation learning approaches based on discrepancy propagation analysis. Under the infinite-horizon setting, the framework leads to the value discrepancy of behavioral cloning in an order of O((1-γ)^{-2}). We also show that the …
Korea Advanced Institute of Science and Technology（韩国科学技术院）；Microsoft Research Asia（微软亚洲研究院）
Behavioral cloning has proven to be effective for learning sequential decision-making policies from expert demonstrations. However, behavioral cloning often suffers from the causal confusion problem where a policy relies on the noticeable effect of expert actions due to the strong correlation but not the cause we desire. This paper presents Object-aware REgularizatiOn (OREO), a simple technique that regularizes an imitation policy in an object-aware manner. Our main idea is to encourage a policy to uniformly attend to all semantic objects, in order to prevent the policy from exploiting nuisanc…
It is difficult to be able to imitate well in unknown states from a small amount of expert data and sampling data. Supervised learning methods such as Behavioral Cloning do not require sampling data, but usually suffer from distribution shift. The methods based on reinforcement learning, such as inverse reinforcement learning and generative adversarial imitation learning (GAIL), can learn from only a few expert data. However, they often need to interact with the environment. Soft Q imitation learning addressed the problems, and it was shown that it could learn efficiently by combining Behavior…
Actor-Critic Pretraining for Proximal Policy Optimization
Institute for Control Engineering of Machine Tools and Manufacturing Units ISW, University of Stuttgart（斯图加特大学机床与制造装置控制工程研究所ISW）；Materials Testing Institute MPA, University of Stuttgart（斯图加特大学材料测试研究所MPA）
Improving Behavioural Cloning with Positive Unlabeled Learning
University College Dublin（都柏林大学）；University College London（伦敦大学学院）；Dublin City University（都柏林城市大学）；Insight SFI Research Centre for Data Analytics（Insight SFI 数据分析研究中心）；MPI for Intelligent Systems（马克斯·普朗克智能系统研究所）
UT Austin（德克萨斯大学奥斯汀分校）；Army Research Laboratory（陆军研究实验室）；Sony AI（索尼人工智能）
Given a dataset of expert agent interactions with an environment of interest, a viable method to extract an effective agent policy is to estimate the maximum likelihood policy indicated by this data. This approach is commonly referred to as behavioral cloning (BC). In this work, we describe a key disadvantage of BC that arises due to the maximum likelihood objective function; namely that BC is mean-seeking with respect to the state-conditional expert action distribution when the learner's policy is represented with a Gaussian. To address this issue, we introduce a modified version of BC, Adver…
Learning complex manipulation tasks in realistic, obstructed environments is a challenging problem due to hard exploration in the presence of obstacles and high-dimensional visual observations. Prior work tackles the exploration problem by integrating motion planning and reinforcement learning. However, the motion planner augmented policy requires access to state information, which is often not available in the real-world settings. To this end, we propose to distill a state-based motion planner augmented policy to a visual control policy via (1) visual behavioral cloning to remove the motion p…
University of Texas at Austin（德克萨斯大学奥斯汀分校）；Tufts University（塔夫茨大学）
Human gaze is known to be an intention-revealing signal in human demonstrations of tasks. In this work, we use gaze cues from human demonstrators to enhance the performance of agents trained via three popular imitation learning methods -- behavioral cloning (BC), behavioral cloning from observation (BCO), and Trajectory-ranked Reward EXtrapolation (T-REX). Based on similarities between the attention of reinforcement learning agents and human gaze, we propose a novel approach for utilizing gaze data in a computationally efficient manner, as part of an auxiliary loss function, which guides a net…
Autonomous Systems Lab (ASL), ETH Zurich（苏黎世联邦理工学院自主系统实验室）；Toyota Research Institute (TRI)（丰田研究所）
Safe autonomous driving requires robust detection of other traffic participants. However, robust does not mean perfect, and safe systems typically minimize missed detections at the expense of a higher false positive rate. This results in conservative and yet potentially dangerous behavior such as avoiding imaginary obstacles. In the context of behavioral cloning, perceptual errors at training time can lead to learning difficulties or wrong policies, as expert demonstrations might be inconsistent with the perceived world state. In this work, we propose a behavioral cloning approach that can saf…
Inverse Reinforcement Learning infers a reward function from expert demonstrations, aiming to encode the behavior and intentions of the expert. Current approaches usually do this with generative and uni-modal models, meaning that they encode a single behavior. In the common setting, where there are various solutions to a problem and the experts show versatile behavior this severely limits the generalization capabilities of these methods. We propose a novel method for Inverse Reinforcement Learning that overcomes these problems by formulating the recovered reward as a sum of iteratively trained…
Collaborating with humans requires rapidly adapting to their individual strengths, weaknesses, and preferences. Unfortunately, most standard multi-agent reinforcement learning techniques, such as self-play (SP) or population play (PP), produce agents that overfit to their training partners and do not generalize well to humans. Alternatively, researchers can collect human data, train a human model using behavioral cloning, and then use that model to train "human-aware" agents ("behavioral cloning play", or BCP). While such an approach can improve the generalization of agents to new human co-pla…
University of Chicago（芝加哥大学）；Toyota Technological Institute at Chicago（芝加哥丰田技术研究所）
Learning from demonstration is widely used as an efficient way for robots to acquire new skills. However, it typically requires that demonstrations provide full access to the state and action sequences. In contrast, learning from observation offers a way to utilize unlabeled demonstrations (e.g., video) to perform imitation learning. One approach to this is behavioral cloning from observation (BCO). The original implementation of BCO proceeds by first learning an inverse dynamics model and then using that model to estimate action labels, thereby reducing the problem to behavioral cloning. Howe…
UC Berkeley（加州大学伯克利分校）；University of Pennsylvania（宾夕法尼亚大学）；Institute for Interdisciplinary Information Sciences, Tsinghua University（清华大学交叉信息研究院）；Shanghai Qi Zhi Institute（上海期智研究院）
Imitation learning trains policies to map from input observations to the actions that an expert would choose. In this setting, distribution shift frequently exacerbates the effect of misattributing expert actions to nuisance correlates among the observed variables. We observe that a common instance of this causal confusion occurs in partially observed settings when expert actions are strongly correlated over time: the imitator learns to cheat by predicting the expert's previous action, rather than the next action. To combat this "copycat problem", we propose an adversarial approach to learn a …
Cornell University（康奈尔大学）；Mitsubishi Electric Research Laboratories（三菱电机研究实验室）
One of the main challenges in peg-in-a-hole (PiH) insertion tasks is in handling the uncertainty in the location of the target hole. In order to address it, high-dimensional sensor inputs from sensor modalities such as vision, force/torque sensing, and proprioception can be combined to learn control policies that are robust to this uncertainty in the target pose. Whereas deep learning has shown success in recognizing objects and making decisions with high-dimensional inputs, the learning procedure might damage the robot when applying directly trial- and-error algorithms on the real system. At …
University of California, Santa Barbara（加州大学圣巴巴拉分校）
Though impressive results have been achieved in visual captioning, the task of generating abstract stories from photo streams is still a little-tapped problem. Different from captions, stories have more expressive language styles and contain many imaginary concepts that do not appear in the images. Thus it poses challenges to behavioral cloning algorithms. Furthermore, due to the limitations of automatic metrics on evaluating story quality, reinforcement learning methods with hand-crafted rewards also face difficulties in gaining an overall performance boost. Therefore, we propose an Adversari…
Johannes Kepler University Linz（林茨约翰内斯·开普勒大学）；Institute of Advanced Research in Artificial Intelligence (IARAI)（人工智能高级研究所（IARAI））；Dynatrace Research（Dynatrace研究院）
The application of Reinforcement Learning (RL) in real world environments can be expensive or risky due to sub-optimal policies during training. In Offline RL, this problem is avoided since interactions with an environment are prohibited. Policies are learned from a given dataset, which solely determines their performance. Despite this fact, how dataset characteristics influence Offline RL algorithms is still hardly investigated. The dataset characteristics are determined by the behavioral policy that samples this dataset. Therefore, we define characteristics of behavioral policies as explorat…
Recent work by Jarrett et al. attempts to frame the problem of offline imitation learning (IL) as one of learning a joint energy-based model, with the hope of out-performing standard behavioral cloning. We suggest that notational issues obscure how the psuedo-state visitation distribution the authors propose to optimize might be disconnected from the policy's $\textit{true}$ state visitation distribution. We further construct natural examples where the parameter coupling advocated by Jarrett et al. leads to inconsistent estimates of the expert's policy, unlike behavioral cloning.
In this work, we present a lightweight pipeline for robust behavioral cloning of a human driver using end-to-end imitation learning. The proposed pipeline was employed to train and deploy three distinct driving behavior models onto a simulated vehicle. The training phase comprised of data collection, balancing, augmentation, preprocessing and training a neural network, following which, the trained model was deployed onto the ego vehicle to predict steering commands based on the feed from an onboard camera. A novel coupled control law was formulated to generate longitudinal control commands on-…
The rapid growth of ride-hailing platforms has created a highly competitive market where businesses struggle to make profits, demanding the need for better operational strategies. However, real-world experiments are risky and expensive for these platforms as they deal with millions of users daily. Thus, a need arises for a simulated environment where they can predict users' reactions to changes in the platform-specific parameters such as trip fares and incentives. Building such a simulation is challenging, as these platforms exist within dynamic environments where thousands of users regularly …
Google Research（谷歌研究院）；Inria（法国国家信息与自动化研究所）；Univ. Grenoble Alpes（格勒诺布尔阿尔卑斯大学）；CNRS（法国国家科学研究中心）；Grenoble INP（格勒诺布尔理工学院）；École Normale Supérieure（巴黎高等师范学院）；PSL Research University（巴黎文理研究大学）
Residual reinforcement learning (RL) has been proposed as a way to solve challenging robotic tasks by adapting control actions from a conventional feedback controller to maximize a reward signal. We extend the residual formulation to learn from visual inputs and sparse rewards using demonstrations. Learning from images, proprioceptive inputs and a sparse task-completion reward relaxes the requirement of accessing full state features, such as object and target positions. In addition, replacing the base controller with a policy learned from demonstrations removes the dependency on a hand-enginee…
Despite its experimental success, Model-based Reinforcement Learning still lacks a complete theoretical understanding. To this end, we analyze the error in the cumulative reward using a contraction approach. We consider both stochastic and deterministic state transitions for continuous (non-discrete) state and action spaces. This approach doesn't require strong assumptions and can recover the typical quadratic error to the horizon. We prove that branched rollouts can reduce this error and are essential for deterministic transitions to have a Bellman contraction. Our analysis of policy mismatch…
Princeton University（普林斯顿大学）；University of California, Berkeley（加州大学伯克利分校）；Stanford University（斯坦福大学）
Imitation learning, followed by reinforcement learning algorithms, is a promising paradigm to solve complex control tasks sample-efficiently. However, learning from demonstrations often suffers from the covariate shift problem, which results in cascading errors of the learned policy. We introduce a notion of conservatively-extrapolated value functions, which provably lead to policies with self-correction. We design an algorithm Value Iteration with Negative Sampling (VINS) that practically learns such value functions with conservative extrapolation. We show that VINS can correct mistakes of th…
Learning from Observation (LfO), also known as Behavioral Cloning, is an approach for building software agents by recording the behavior of an expert (human or artificial) and using the recorded data to generate the required behavior. jLOAF is a platform that uses Case-Based Reasoning to achieve LfO. In this paper we interface jLOAF with the popular OpenAI Gym environment. Our experimental results show how our approach can be used to provide a baseline for comparison in this domain, as well as identify the strengths and weaknesses when dealing with environmental complexity.
In this work, we argue for the importance of an online evaluation budget for a reliable comparison of deep offline RL algorithms. First, we delineate that the online evaluation budget is problem-dependent, where some problems allow for less but others for more. And second, we demonstrate that the preference between algorithms is budget-dependent across a diverse range of decision-making domains such as Robotics, Finance, and Energy Management. Following the points above, we suggest reporting the performance of deep offline RL algorithms under varying online evaluation budgets. To facilitate th…
