Goal-Conditioned Reinforcement Learning (GCRL) can enable agents to spontaneously set diverse goals to learn a set of skills. Despite the excellent works proposed in various fields, reaching distant goals in temporally extended tasks remains a challenge for GCRL. Current works tackled this problem by leveraging planning algorithms to plan intermediate subgoals to augment GCRL. Their methods need two crucial requirements: (i) a state representation space to search valid subgoals, and (ii) a distance function to measure the reachability of subgoals. However, they struggle to scale to high-dimensional state space due to their non-compact representations. Moreover, they cannot collect high-quality training data through standard GC policies, which results in an inaccurate distance function. Both affect the efficiency and performance of planning and policy learning. In the paper, we propose a goal-conditioned RL algorithm combined with Disentanglement-based Reachability Planning (REPlan) to solve temporally extended tasks. In REPlan, a Disentangled Representation Module (DRM) is proposed to learn compact representations which disentangle robot poses and object positions from high-dimensional observations in a self-supervised manner. A simple REachability discrimination Module (REM) is also designed to determine the temporal distance of subgoals. Moreover, REM computes intrinsic bonuses to encourage the collection of novel states for training. We evaluate our REPlan in three vision-based simulation tasks and one real-world task. The experiments demonstrate that our REPlan significantly outperforms the prior state-of-the-art methods in solving temporally extended tasks.
相關內容
Continual Learning has been challenging, especially when dealing with unsupervised scenarios such as Unsupervised Online General Continual Learning (UOGCL), where the learning agent has no prior knowledge of class boundaries or task change information. While previous research has focused on reducing forgetting in supervised setups, recent studies have shown that self-supervised learners are more resilient to forgetting. This paper proposes a novel approach that enhances memory usage for contrastive learning in UOGCL by defining and using stream-dependent data augmentations together with some implementation tricks. Our proposed method is simple yet effective, achieves state-of-the-art results compared to other unsupervised approaches in all considered setups, and reduces the gap between supervised and unsupervised continual learning. Our domain-aware augmentation procedure can be adapted to other replay-based methods, making it a promising strategy for continual learning.
Deep Reinforcement Learning (DRL) has achieved remarkable success in sequential decision-making problems. However, existing DRL agents make decisions in an opaque fashion, hindering the user from establishing trust and scrutinizing weaknesses of the agents. While recent research has developed Interpretable Policy Extraction (IPE) methods for explaining how an agent takes actions, their explanations are often inconsistent with the agent's behavior and thus, frequently fail to explain. To tackle this issue, we propose a novel method, Fidelity-Induced Policy Extraction (FIPE). Specifically, we start by analyzing the optimization mechanism of existing IPE methods, elaborating on the issue of ignoring consistency while increasing cumulative rewards. We then design a fidelity-induced mechanism by integrate a fidelity measurement into the reinforcement learning feedback. We conduct experiments in the complex control environment of StarCraft II, an arena typically avoided by current IPE methods. The experiment results demonstrate that FIPE outperforms the baselines in terms of interaction performance and consistency, meanwhile easy to understand.
For Prognostics and Health Management (PHM) of Lithium-ion (Li-ion) batteries, many models have been established to characterize their degradation process. The existing empirical or physical models can reveal important information regarding the degradation dynamics. However, there are no general and flexible methods to fuse the information represented by those models. Physics-Informed Neural Network (PINN) is an efficient tool to fuse empirical or physical dynamic models with data-driven models. To take full advantage of various information sources, we propose a model fusion scheme based on PINN. It is implemented by developing a semi-empirical semi-physical Partial Differential Equation (PDE) to model the degradation dynamics of Li-ion batteries. When there is little prior knowledge about the dynamics, we leverage the data-driven Deep Hidden Physics Model (DeepHPM) to discover the underlying governing dynamic models. The uncovered dynamics information is then fused with that mined by the surrogate neural network in the PINN framework. Moreover, an uncertainty-based adaptive weighting method is employed to balance the multiple learning tasks when training the PINN. The proposed methods are verified on a public dataset of Li-ion Phosphate (LFP)/graphite batteries.
Class-Incremental learning (CIL) is the ability of artificial agents to accommodate new classes as they appear in a stream. It is particularly interesting in evolving environments where agents have limited access to memory and computational resources. The main challenge of class-incremental learning is catastrophic forgetting, the inability of neural networks to retain past knowledge when learning a new one. Unfortunately, most existing class-incremental object detectors are applied to two-stage algorithms such as Faster-RCNN and rely on rehearsal memory to retain past knowledge. We believe that the current benchmarks are not realistic, and more effort should be dedicated to anchor-free and rehearsal-free object detection. In this context, we propose MultIOD, a class-incremental object detector based on CenterNet. Our main contributions are: (1) we propose a multihead feature pyramid and multihead detection architecture to efficiently separate class representations, (2) we employ transfer learning between classes learned initially and those learned incrementally to tackle catastrophic forgetting, and (3) we use a class-wise non-max-suppression as a post-processing technique to remove redundant boxes. Without bells and whistles, our method outperforms a range of state-of-the-art methods on two Pascal VOC datasets.
The mixing of two or more languages is called Code-Mixing (CM). CM is a social norm in multilingual societies. Neural Language Models (NLMs) like transformers have been very effective on many NLP tasks. However, NLM for CM is an under-explored area. Though transformers are capable and powerful, they cannot always encode positional/sequential information since they are non-recurrent. Therefore, to enrich word information and incorporate positional information, positional encoding is defined. We hypothesize that Switching Points (SPs), i.e., junctions in the text where the language switches (L1 -> L2 or L2-> L1), pose a challenge for CM Language Models (LMs), and hence give special emphasis to switching points in the modeling process. We experiment with several positional encoding mechanisms and show that rotatory positional encodings along with switching point information yield the best results. We introduce CONFLATOR: a neural language modeling approach for code-mixed languages. CONFLATOR tries to learn to emphasize switching points using smarter positional encoding, both at unigram and bigram levels. CONFLATOR outperforms the state-of-the-art on two tasks based on code-mixed Hindi and English (Hinglish): (i) sentiment analysis and (ii) machine translation.
Recently Transformer and Convolution neural network (CNN) based models have shown promising results in EEG signal processing. Transformer models can capture the global dependencies in EEG signals through a self-attention mechanism, while CNN models can capture local features such as sawtooth waves. In this work, we propose an end-to-end neural epilepsy detection model, EENED, that combines CNN and Transformer. Specifically, by introducing the convolution module into the Transformer encoder, EENED can learn the time-dependent relationship of the patient's EEG signal features and notice local EEG abnormal mutations closely related to epilepsy, such as the appearance of spikes and the sprinkling of sharp and slow waves. Our proposed framework combines the ability of Transformer and CNN to capture different scale features of EEG signals and holds promise for improving the accuracy and reliability of epilepsy detection. Our source code will be released soon on GitHub.
Face Anti-Spoofing (FAS) aims to detect malicious attempts to invade a face recognition system by presenting spoofed faces. State-of-the-art FAS techniques predominantly rely on deep learning models but their cross-domain generalization capabilities are often hindered by the domain shift problem, which arises due to different distributions between training and testing data. In this study, we develop a generalized FAS method under the Efficient Parameter Transfer Learning (EPTL) paradigm, where we adapt the pre-trained Vision Transformer models for the FAS task. During training, the adapter modules are inserted into the pre-trained ViT model, and the adapters are updated while other pre-trained parameters remain fixed. We find the limitations of previous vanilla adapters in that they are based on linear layers, which lack a spoofing-aware inductive bias and thus restrict the cross-domain generalization. To address this limitation and achieve cross-domain generalized FAS, we propose a novel Statistical Adapter (S-Adapter) that gathers local discriminative and statistical information from localized token histograms. To further improve the generalization of the statistical tokens, we propose a novel Token Style Regularization (TSR), which aims to reduce domain style variance by regularizing Gram matrices extracted from tokens across different domains. Our experimental results demonstrate that our proposed S-Adapter and TSR provide significant benefits in both zero-shot and few-shot cross-domain testing, outperforming state-of-the-art methods on several benchmark tests. We will release the source code upon acceptance.
In Multi-Label Text Classification (MLTC), one sample can belong to more than one class. It is observed that most MLTC tasks, there are dependencies or correlations among labels. Existing methods tend to ignore the relationship among labels. In this paper, a graph attention network-based model is proposed to capture the attentive dependency structure among the labels. The graph attention network uses a feature matrix and a correlation matrix to capture and explore the crucial dependencies between the labels and generate classifiers for the task. The generated classifiers are applied to sentence feature vectors obtained from the text feature extraction network (BiLSTM) to enable end-to-end training. Attention allows the system to assign different weights to neighbor nodes per label, thus allowing it to learn the dependencies among labels implicitly. The results of the proposed model are validated on five real-world MLTC datasets. The proposed model achieves similar or better performance compared to the previous state-of-the-art models.
Recommender systems are widely used in big information-based companies such as Google, Twitter, LinkedIn, and Netflix. A recommender system deals with the problem of information overload by filtering important information fragments according to users' preferences. In light of the increasing success of deep learning, recent studies have proved the benefits of using deep learning in various recommendation tasks. However, most proposed techniques only aim to target individuals, which cannot be efficiently applied in group recommendation. In this paper, we propose a deep learning architecture to solve the group recommendation problem. On the one hand, as different individual preferences in a group necessitate preference trade-offs in making group recommendations, it is essential that the recommendation model can discover substitutes among user behaviors. On the other hand, it has been observed that a user as an individual and as a group member behaves differently. To tackle such problems, we propose using an attention mechanism to capture the impact of each user in a group. Specifically, our model automatically learns the influence weight of each user in a group and recommends items to the group based on its members' weighted preferences. We conduct extensive experiments on four datasets. Our model significantly outperforms baseline methods and shows promising results in applying deep learning to the group recommendation problem.
We propose a novel single shot object detection network named Detection with Enriched Semantics (DES). Our motivation is to enrich the semantics of object detection features within a typical deep detector, by a semantic segmentation branch and a global activation module. The segmentation branch is supervised by weak segmentation ground-truth, i.e., no extra annotation is required. In conjunction with that, we employ a global activation module which learns relationship between channels and object classes in a self-supervised manner. Comprehensive experimental results on both PASCAL VOC and MS COCO detection datasets demonstrate the effectiveness of the proposed method. In particular, with a VGG16 based DES, we achieve an mAP of 81.7 on VOC2007 test and an mAP of 32.8 on COCO test-dev with an inference speed of 31.5 milliseconds per image on a Titan Xp GPU. With a lower resolution version, we achieve an mAP of 79.7 on VOC2007 with an inference speed of 13.0 milliseconds per image.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191