This work studies the problem of image-goal navigation, which entails guiding robots with noisy sensors and controls through real crowded environments. Recent fruitful approaches rely on deep reinforcement learning and learn navigation policies in simulation environments that are much simpler in complexity than real environments. Directly transferring these trained policies to real environments can be extremely challenging or even dangerous. We tackle this problem with a hierarchical navigation method composed of four decoupled modules. The first module maintains an obstacle map during robot navigation. The second one predicts a long-term goal on the real-time map periodically. The third one plans collision-free command sets for navigating to long-term goals, while the final module stops the robot properly near the goal image. The four modules are developed separately to suit the image-goal navigation in real crowded scenarios. In addition, the hierarchical decomposition decouples the learning of navigation goal planning, collision avoidance and navigation ending prediction, which cuts down the search space during navigation training and helps improve the generalization to previously unseen real scenes. We evaluate the method in both a simulator and the real world with a mobile robot. The results show that our method outperforms several navigation baselines and can successfully achieve navigation tasks in these scenarios.
相關內容
We introduce a new constrained optimization method for policy gradient reinforcement learning, which uses two trust regions to regulate each policy update. In addition to using the proximity of one single old policy as the first trust region as done by prior works, we propose to form a second trust region through the construction of another virtual policy that represents a wide range of past policies. We then enforce the new policy to stay closer to the virtual policy, which is beneficial in case the old policy performs badly. More importantly, we propose a mechanism to automatically build the virtual policy from a memory buffer of past policies, providing a new capability for dynamically selecting appropriate trust regions during the optimization process. Our proposed method, dubbed as Memory-Constrained Policy Optimization (MCPO), is examined on a diverse suite of environments including robotic locomotion control, navigation with sparse rewards and Atari games, consistently demonstrating competitive performance against recent on-policy constrained policy gradient methods.
Embodied AI is a recent research area that aims at creating intelligent agents that can move and operate inside an environment. Existing approaches in this field demand the agents to act in completely new and unexplored scenes. However, this setting is far from realistic use cases that instead require executing multiple tasks in the same environment. Even if the environment changes over time, the agent could still count on its global knowledge about the scene while trying to adapt its internal representation to the current state of the environment. To make a step towards this setting, we propose Spot the Difference: a novel task for Embodied AI where the agent has access to an outdated map of the environment and needs to recover the correct layout in a fixed time budget. To this end, we collect a new dataset of occupancy maps starting from existing datasets of 3D spaces and generating a number of possible layouts for a single environment. This dataset can be employed in the popular Habitat simulator and is fully compliant with existing methods that employ reconstructed occupancy maps during navigation. Furthermore, we propose an exploration policy that can take advantage of previous knowledge of the environment and identify changes in the scene faster and more effectively than existing agents. Experimental results show that the proposed architecture outperforms existing state-of-the-art models for exploration on this new setting.
This paper identifies and addresses a serious design bias of existing salient object detection (SOD) datasets, which unrealistically assume that each image should contain at least one clear and uncluttered salient object. This design bias has led to a saturation in performance for state-of-the-art SOD models when evaluated on existing datasets. However, these models are still far from satisfactory when applied to real-world scenes. Based on our analyses, we propose a new high-quality dataset and update the previous saliency benchmark. Specifically, our dataset, called Salient Objects in Clutter~\textbf{(SOC)}, includes images with both salient and non-salient objects from several common object categories. In addition to object category annotations, each salient image is accompanied by attributes that reflect common challenges in common scenes, which can help provide deeper insight into the SOD problem. Further, with a given saliency encoder, e.g., the backbone network, existing saliency models are designed to achieve mapping from the training image set to the training ground-truth set. We, therefore, argue that improving the dataset can yield higher performance gains than focusing only on the decoder design. With this in mind, we investigate several dataset-enhancement strategies, including label smoothing to implicitly emphasize salient boundaries, random image augmentation to adapt saliency models to various scenarios, and self-supervised learning as a regularization strategy to learn from small datasets. Our extensive results demonstrate the effectiveness of these tricks. We also provide a comprehensive benchmark for SOD, which can be found in our repository: //github.com/DengPingFan/SODBenchmark.
Driving safely requires multiple capabilities from human and intelligent agents, such as the generalizability to unseen environments, the safety awareness of the surrounding traffic, and the decision-making in complex multi-agent settings. Despite the great success of Reinforcement Learning (RL), most of the RL research works investigate each capability separately due to the lack of integrated environments. In this work, we develop a new driving simulation platform called MetaDrive to support the research of generalizable reinforcement learning algorithms for machine autonomy. MetaDrive is highly compositional, which can generate an infinite number of diverse driving scenarios from both the procedural generation and the real data importing. Based on MetaDrive, we construct a variety of RL tasks and baselines in both single-agent and multi-agent settings, including benchmarking generalizability across unseen scenes, safe exploration, and learning multi-agent traffic. The generalization experiments conducted on both procedurally generated scenarios and real-world scenarios show that increasing the diversity and the size of the training set leads to the improvement of the generalizability of the RL agents. We further evaluate various safe reinforcement learning and multi-agent reinforcement learning algorithms in MetaDrive environments and provide the benchmarks. Source code, documentation, and demo video are available at //metadriverse.github.io/metadrive . More research projects based on MetaDrive simulator are listed at //metadriverse.github.io
In recent years, malware detection has become an active research topic in the area of Internet of Things (IoT) security. The principle is to exploit knowledge from large quantities of continuously generated malware. Existing algorithms practice available malware features for IoT devices and lack real-time prediction behaviors. More research is thus required on malware detection to cope with real-time misclassification of the input IoT data. Motivated by this, in this paper we propose an adversarial self-supervised architecture for detecting malware in IoT networks, SETTI, considering samples of IoT network traffic that may not be labeled. In the SETTI architecture, we design three self-supervised attack techniques, namely Self-MDS, GSelf-MDS and ASelf-MDS. The Self-MDS method considers the IoT input data and the adversarial sample generation in real-time. The GSelf-MDS builds a generative adversarial network model to generate adversarial samples in the self-supervised structure. Finally, ASelf-MDS utilizes three well-known perturbation sample techniques to develop adversarial malware and inject it over the self-supervised architecture. Also, we apply a defence method to mitigate these attacks, namely adversarial self-supervised training to protect the malware detection architecture against injecting the malicious samples. To validate the attack and defence algorithms, we conduct experiments on two recent IoT datasets: IoT23 and NBIoT. Comparison of the results shows that in the IoT23 dataset, the Self-MDS method has the most damaging consequences from the attacker's point of view by reducing the accuracy rate from 98% to 74%. In the NBIoT dataset, the ASelf-MDS method is the most devastating algorithm that can plunge the accuracy rate from 98% to 77%.
The best neural architecture for a given machine learning problem depends on many factors: not only the complexity and structure of the dataset, but also on resource constraints including latency, compute, energy consumption, etc. Neural architecture search (NAS) for tabular datasets is an important but under-explored problem. Previous NAS algorithms designed for image search spaces incorporate resource constraints directly into the reinforcement learning rewards. In this paper, we argue that search spaces for tabular NAS pose considerable challenges for these existing reward-shaping methods, and propose a new reinforcement learning (RL) controller to address these challenges. Motivated by rejection sampling, when we sample candidate architectures during a search, we immediately discard any architecture that violates our resource constraints. We use a Monte-Carlo-based correction to our RL policy gradient update to account for this extra filtering step. Results on several tabular datasets show TabNAS, the proposed approach, efficiently finds high-quality models that satisfy the given resource constraints.
Graph Neural Networks (GNNs), which generalize deep neural networks to graph-structured data, have drawn considerable attention and achieved state-of-the-art performance in numerous graph related tasks. However, existing GNN models mainly focus on designing graph convolution operations. The graph pooling (or downsampling) operations, that play an important role in learning hierarchical representations, are usually overlooked. In this paper, we propose a novel graph pooling operator, called Hierarchical Graph Pooling with Structure Learning (HGP-SL), which can be integrated into various graph neural network architectures. HGP-SL incorporates graph pooling and structure learning into a unified module to generate hierarchical representations of graphs. More specifically, the graph pooling operation adaptively selects a subset of nodes to form an induced subgraph for the subsequent layers. To preserve the integrity of graph's topological information, we further introduce a structure learning mechanism to learn a refined graph structure for the pooled graph at each layer. By combining HGP-SL operator with graph neural networks, we perform graph level representation learning with focus on graph classification task. Experimental results on six widely used benchmarks demonstrate the effectiveness of our proposed model.
The goal of few-shot learning is to learn a classifier that generalizes well even when trained with a limited number of training instances per class. The recently introduced meta-learning approaches tackle this problem by learning a generic classifier across a large number of multiclass classification tasks and generalizing the model to a new task. Yet, even with such meta-learning, the low-data problem in the novel classification task still remains. In this paper, we propose Transductive Propagation Network (TPN), a novel meta-learning framework for transductive inference that classifies the entire test set at once to alleviate the low-data problem. Specifically, we propose to learn to propagate labels from labeled instances to unlabeled test instances, by learning a graph construction module that exploits the manifold structure in the data. TPN jointly learns both the parameters of feature embedding and the graph construction in an end-to-end manner. We validate TPN on multiple benchmark datasets, on which it largely outperforms existing few-shot learning approaches and achieves the state-of-the-art results.
We introduce a generic framework that reduces the computational cost of object detection while retaining accuracy for scenarios where objects with varied sizes appear in high resolution images. Detection progresses in a coarse-to-fine manner, first on a down-sampled version of the image and then on a sequence of higher resolution regions identified as likely to improve the detection accuracy. Built upon reinforcement learning, our approach consists of a model (R-net) that uses coarse detection results to predict the potential accuracy gain for analyzing a region at a higher resolution and another model (Q-net) that sequentially selects regions to zoom in. Experiments on the Caltech Pedestrians dataset show that our approach reduces the number of processed pixels by over 50% without a drop in detection accuracy. The merits of our approach become more significant on a high resolution test set collected from YFCC100M dataset, where our approach maintains high detection performance while reducing the number of processed pixels by about 70% and the detection time by over 50%.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191