Reward design is a fundamental problem in reinforcement learning (RL). A misspecified or poorly designed reward can result in low sample efficiency and undesired behaviors. In this paper, we propose the idea of programmatic reward design, i.e. using programs to specify the reward functions in RL environments. Programs allow human engineers to express sub-goals and complex task scenarios in a structured and interpretable way. The challenge of programmatic reward design, however, is that while humans can provide the high-level structures, properly setting the low-level details, such as the right amount of reward for a specific sub-task, remains difficult. A major contribution of this paper is a probabilistic framework that can infer the best candidate programmatic reward function from expert demonstrations. Inspired by recent generative-adversarial approaches, our framework searches for the most likely programmatic reward function under which the optimally generated trajectories cannot be differentiated from the demonstrated trajectories. Experimental results show that programmatic reward functionslearned using this framework can significantly outperform those learned using existing reward learning algo-rithms, and enable RL agents to achieve state-of-the-artperformance on highly complex tasks.
相關內容
Continuous-time (CT) models have shown an improved sample efficiency during learning and enable ODE analysis methods for enhanced interpretability compared to discrete-time (DT) models. Even with numerous recent developments, the multifaceted CT state-space model identification problem remains to be solved in full, considering common experimental aspects such as the presence of external inputs, measurement noise, and latent states. This paper presents a novel estimation method that includes these aspects and that is able to obtain state-of-the-art results on multiple benchmarks where a small fully connected neural network describes the CT dynamics. The novel estimation method called the subspace encoder approach ascertains these results by altering the well-known simulation loss to include short subsections instead, by using an encoder function and a state-derivative normalization term to obtain a computationally feasible and stable optimization problem. This encoder function estimates the initial states of each considered subsection. We prove that the existence of the encoder function has the necessary condition of a Lipschitz continuous state-derivative utilizing established properties of ODEs.
Human visual attention is a complex phenomenon that has been studied for decades. Within it, the particular problem of scanpath prediction poses a challenge, particularly due to the inter- and intra-observer variability, among other reasons. Besides, most existing approaches to scanpath prediction have focused on optimizing the prediction of a gaze point given the previous ones. In this work, we present a probabilistic time-evolving approach to scanpath prediction, based on Bayesian deep learning. We optimize our model using a novel spatio-temporal loss function based on a combination of Kullback-Leibler divergence and dynamic time warping, jointly considering the spatial and temporal dimensions of scanpaths. Our scanpath prediction framework yields results that outperform those of current state-of-the-art approaches, and are almost on par with the human baseline, suggesting that our model is able to generate scanpaths whose behavior closely resembles those of the real ones.
Feature quality has an impactful effect on recommendation performance. Thereby, feature selection is a critical process in developing deep learning-based recommender systems. Most existing deep recommender systems, however, focus on designing sophisticated neural networks, while neglecting the feature selection process. Typically, they just feed all possible features into their proposed deep architectures, or select important features manually by human experts. The former leads to non-trivial embedding parameters and extra inference time, while the latter requires plenty of expert knowledge and human labor effort. In this work, we propose an AutoML framework that can adaptively select the essential feature fields in an automatic manner. Specifically, we first design a differentiable controller network, which is capable of automatically adjusting the probability of selecting a particular feature field; then, only selected feature fields are utilized to retrain the deep recommendation model. Extensive experiments on three benchmark datasets demonstrate the effectiveness of our framework. We conduct further experiments to investigate its properties, including the transferability, key components, and parameter sensitivity.
This work proposes an extension to Structured Grammatical Evolution (SGE) called Co-evolutionary Probabilistic Structured Grammatical Evolution (Co-PSGE). In Co-PSGE each individual in the population is composed by a grammar and a genotype, which is a list of dynamic lists, each corresponding to a non-terminal of the grammar containing real numbers that correspond to the probability of choosing a derivation rule. Each individual uses its own grammar to map the genotype into a program. During the evolutionary process, both the grammar and the genotype are subject to variation operators. The performance of the proposed approach is compared to 3 different methods, namely, Grammatical Evolution (GE), Probabilistic Grammatical Evolution (PGE), and SGE on four different benchmark problems. The results show the effectiveness of the approach since Co-PSGE is able to outperform all the methods with statistically significant differences in the majority of the problems.
Bayesian policy reuse (BPR) is a general policy transfer framework for selecting a source policy from an offline library by inferring the task belief based on some observation signals and a trained observation model. In this paper, we propose an improved BPR method to achieve more efficient policy transfer in deep reinforcement learning (DRL). First, most BPR algorithms use the episodic return as the observation signal that contains limited information and cannot be obtained until the end of an episode. Instead, we employ the state transition sample, which is informative and instantaneous, as the observation signal for faster and more accurate task inference. Second, BPR algorithms usually require numerous samples to estimate the probability distribution of the tabular-based observation model, which may be expensive and even infeasible to learn and maintain, especially when using the state transition sample as the signal. Hence, we propose a scalable observation model based on fitting state transition functions of source tasks from only a small number of samples, which can generalize to any signals observed in the target task. Moreover, we extend the offline-mode BPR to the continual learning setting by expanding the scalable observation model in a plug-and-play fashion, which can avoid negative transfer when faced with new unknown tasks. Experimental results show that our method can consistently facilitate faster and more efficient policy transfer.
We present a data-efficient framework for solving sequential decision-making problems which exploits the combination of reinforcement learning (RL) and latent variable generative models. The framework, called GenRL, trains deep policies by introducing an action latent variable such that the feed-forward policy search can be divided into two parts: (i) training a sub-policy that outputs a distribution over the action latent variable given a state of the system, and (ii) unsupervised training of a generative model that outputs a sequence of motor actions conditioned on the latent action variable. GenRL enables safe exploration and alleviates the data-inefficiency problem as it exploits prior knowledge about valid sequences of motor actions. Moreover, we provide a set of measures for evaluation of generative models such that we are able to predict the performance of the RL policy training prior to the actual training on a physical robot. We experimentally determine the characteristics of generative models that have most influence on the performance of the final policy training on two robotics tasks: shooting a hockey puck and throwing a basketball. Furthermore, we empirically demonstrate that GenRL is the only method which can safely and efficiently solve the robotics tasks compared to two state-of-the-art RL methods.
Deep learning techniques for point clouds have achieved strong performance on a range of 3D vision tasks. However, it is costly to annotate large-scale point sets, making it critical to learn generalizable representations that can transfer well across different point sets. In this paper, we study a new problem of 3D Domain Generalization (3DDG) with the goal to generalize the model to other unseen domains of point clouds without any access to them in the training process. It is a challenging problem due to the substantial geometry shift from simulated to real data, such that most existing 3D models underperform due to overfitting the complete geometries in the source domain. We propose to tackle this problem via MetaSets, which meta-learns point cloud representations from a group of classification tasks on carefully-designed transformed point sets containing specific geometry priors. The learned representations are more generalizable to various unseen domains of different geometries. We design two benchmarks for Sim-to-Real transfer of 3D point clouds. Experimental results show that MetaSets outperforms existing 3D deep learning methods by large margins.
Humans and animals have the ability to continually acquire, fine-tune, and transfer knowledge and skills throughout their lifespan. This ability, referred to as lifelong learning, is mediated by a rich set of neurocognitive mechanisms that together contribute to the development and specialization of our sensorimotor skills as well as to long-term memory consolidation and retrieval. Consequently, lifelong learning capabilities are crucial for autonomous agents interacting in the real world and processing continuous streams of information. However, lifelong learning remains a long-standing challenge for machine learning and neural network models since the continual acquisition of incrementally available information from non-stationary data distributions generally leads to catastrophic forgetting or interference. This limitation represents a major drawback for state-of-the-art deep neural network models that typically learn representations from stationary batches of training data, thus without accounting for situations in which information becomes incrementally available over time. In this review, we critically summarize the main challenges linked to lifelong learning for artificial learning systems and compare existing neural network approaches that alleviate, to different extents, catastrophic forgetting. We discuss well-established and emerging research motivated by lifelong learning factors in biological systems such as structural plasticity, memory replay, curriculum and transfer learning, intrinsic motivation, and multisensory integration.
High spectral dimensionality and the shortage of annotations make hyperspectral image (HSI) classification a challenging problem. Recent studies suggest that convolutional neural networks can learn discriminative spatial features, which play a paramount role in HSI interpretation. However, most of these methods ignore the distinctive spectral-spatial characteristic of hyperspectral data. In addition, a large amount of unlabeled data remains an unexploited gold mine for efficient data use. Therefore, we proposed an integration of generative adversarial networks (GANs) and probabilistic graphical models for HSI classification. Specifically, we used a spectral-spatial generator and a discriminator to identify land cover categories of hyperspectral cubes. Moreover, to take advantage of a large amount of unlabeled data, we adopted a conditional random field to refine the preliminary classification results generated by GANs. Experimental results obtained using two commonly studied datasets demonstrate that the proposed framework achieved encouraging classification accuracy using a small number of data for training.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191