Chenxi Sun

Chenxi Sun, Hongyan Li, Yaliang Li et al.

This work summarizes two ways to accomplish Time-Series (TS) tasks in today's Large Language Model (LLM) context: LLM-for-TS (model-centric) designs and trains a fundamental large model, or fine-tunes a pre-trained LLM for TS data; TS-for-LLM (data-centric) converts TS into a model-friendly representation to enable the pre-trained LLM to handle TS data. Given the lack of data, limited resources, semantic context requirements, and so on, this work focuses on TS-for-LLM, where we aim to activate LLM's ability for TS data by designing a TS embedding method suitable for LLM. The proposed method is named TEST. It first tokenizes TS, builds an encoder to embed TS via instance-wise, feature-wise, and text-prototype-aligned contrast, where the TS embedding space is aligned to LLM embedding layer space, then creates soft prompts to make LLM more open to that embeddings, and finally implements TS tasks using the frozen LLM. We also demonstrate the feasibility of TS-for-LLM through theory and experiments. Experiments are carried out on TS classification, forecasting, and representation tasks using eight frozen LLMs with various structures and sizes. The results show that the pre-trained LLM with TEST strategy can achieve better or comparable performance than today's SOTA TS models and offer benefits for few-shot and generalization. By treating LLM as the pattern machine, TEST can endow LLM's ability to process TS data without compromising language ability. We hope that this study will serve as a foundation for future work to support TS+LLM progress.

Fanheng Kong, Jingyuan Zhang, Yang Yue et al.

The emergence of Large Language Models (LLMs) has catalyzed a paradigm shift in programming, giving rise to "vibe coding", where users can build complete projects and even control computers using natural language instructions. This paradigm has driven automated webpage development, but it introduces a new requirement about how to automatically verify whether the web functionalities are reliably implemented. Existing works struggle to adapt, relying on static visual similarity or predefined checklists that constrain their utility in open-ended environments. Furthermore, they overlook a vital aspect of software quality, namely latent logical constraints. To address these gaps, we introduce WebTestBench, a benchmark for evaluating end-to-end automated web testing. WebTestBench encompasses comprehensive dimensions across diverse web application categories. We decompose the testing process into two cascaded sub-tasks, checklist generation and defect detection, and propose WebTester, a baseline framework for this task. Evaluating popular LLMs with WebTester reveals severe challenges, including insufficient test completeness, detection bottlenecks, and long-horizon interaction unreliability. These findings expose a substantial gap between current computer-use agent capabilities and industrial-grade deployment demands. We hope that WebTestBench provides valuable insights and guidance for advancing end-to-end automated web testing. Our dataset and code are available at https://github.com/friedrichor/WebTestBench.

Qi Wang, Hongzhi Zhang, Jia Fu et al.

Despite the proliferation of powerful agentic models, the lack of critical post-training details hinders the development of strong counterparts in the open-source community. In this study, we present a comprehensive and fully open-source pipeline for training a high-performance agentic model for interacting with external tools and environments, named Klear-Qwen3-AgentForge, starting from the Qwen3-8B base model. We design effective supervised fine-tuning (SFT) with synthetic data followed by multi-turn reinforcement learning (RL) to unlock the potential for multiple diverse agentic tasks. We perform exclusive experiments on various agentic benchmarks in both tool use and coding domains. Klear-Qwen3-AgentForge-8B achieves state-of-the-art performance among LLMs of similar size and remains competitive with significantly larger models.

Chenxi Sun, Hongyan Li, Moxian Song et al.

Continuous diagnosis and prognosis are essential for intensive care patients. It can provide more opportunities for timely treatment and rational resource allocation, especially for sepsis, a main cause of death in ICU, and COVID-19, a new worldwide epidemic. Although deep learning methods have shown their great superiority in many medical tasks, they tend to catastrophically forget, over fit, and get results too late when performing diagnosis and prognosis in the continuous mode. In this work, we summarized the three requirements of this task, proposed a new concept, continuous classification of time series (CCTS), and designed a novel model training method, restricted update strategy of neural networks (RU). In the context of continuous prognosis, our method outperformed all baselines and achieved the average accuracy of 90%, 97%, and 85% on sepsis prognosis, COVID-19 mortality prediction, and eight diseases classification. Superiorly, our method can also endow deep learning with interpretability, having the potential to explore disease mechanisms and provide a new horizon for medical research. We have achieved disease staging for sepsis and COVID-19, discovering four stages and three stages with their typical biomarkers respectively. Further, our method is a data-agnostic and model-agnostic plug-in, it can be used to continuously prognose other diseases with staging and even implement CCTS in other fields.

Chenxi Sun, Moxian Song, Derun Can et al.

In the real world, the class of a time series is usually labeled at the final time, but many applications require to classify time series at every time point. e.g. the outcome of a critical patient is only determined at the end, but he should be diagnosed at all times for timely treatment. Thus, we propose a new concept: Continuous Classification of Time Series (CCTS). It requires the model to learn data in different time stages. But the time series evolves dynamically, leading to different data distributions. When a model learns multi-distribution, it always forgets or overfits. We suggest that meaningful learning scheduling is potential due to an interesting observation: Measured by confidence, the process of model learning multiple distributions is similar to the process of human learning multiple knowledge. Thus, we propose a novel Confidence-guided method for CCTS (C3TS). It can imitate the alternating human confidence described by the Dunning-Kruger Effect. We define the objective- confidence to arrange data, and the self-confidence to control the learning duration. Experiments on four real-world datasets show that C3TS is more accurate than all baselines for CCTS.

Yahui Liu, Yang Yue, Jingyuan Zhang et al.

Recent efforts on Diffusion Mixture-of-Experts (MoE) models have primarily focused on developing more sophisticated routing mechanisms. However, we observe that the underlying architectural configuration space remains markedly under-explored. Inspired by the MoE design paradigms established in large language models (LLMs), we identify a set of crucial architectural factors for building effective Diffusion MoE models--including DeepSeek-style expert modules, alternative intermediate widths, varying expert counts, and enhanced attention positional encodings. Our systematic study reveals that carefully tuning these configurations is essential for unlocking the full potential of Diffusion MoE models, often yielding gains that exceed those achieved by routing innovations alone. Through extensive experiments, we present novel architectures that can be efficiently applied to both latent and pixel-space diffusion frameworks, which provide a practical and efficient training recipe that enables Diffusion MoE models to surpass strong baselines while using equal or fewer activated parameters. All code and models are publicly available at: https://github.com/yhlleo/EfficientMoE.

Zhongheng Jiang, Yuechao Zhao, Donglin Xie et al.

Medical time-series data captures the dynamic progression of patient conditions, playing a vital role in modern clinical decision support systems. However, real-world clinical data is highly heterogeneous and inconsistently formatted. Furthermore, existing machine learning tools often have steep learning curves and fragmented workflows. Consequently, a significant gap remains between cutting-edge AI technologies and clinical application. To address this, we introduce mtslearn, an end-to-end integrated toolkit specifically designed for medical time-series data. First, the framework provides a unified data interface that automates the parsing and alignment of wide, long, and flat data formats. This design significantly reduces data cleaning overhead. Building on this, mtslearn provides a complete pipeline from data reading and feature engineering to model training and result visualization. Furthermore, it offers flexible interfaces for custom algorithms. Through a modular design, mtslearn simplifies complex data engineering tasks into a few lines of code. This significantly lowers the barrier to entry for clinicians with limited programming experience, empowering them to focus more on exploring medical hypotheses and accelerating the translation of advanced algorithms into real-world clinical practice. mtslearn is publicly available at https://github.com/PKUDigitalHealth/mtslearn.

Xingguang Ji, Yahui Liu, Qi Wang et al.

We introduce our Leanabell-Prover-V2, a 7B large language models (LLMs) that can produce formal theorem proofs in Lean 4, with verifier-integrated Long Chain-of-Thoughts (CoT). Following our previous work Leanabell-Prover-V1, we continual to choose to posttrain existing strong prover models for further performance improvement. In our V2 version, we mainly upgrade the Reinforcement Learning (RL) with feedback provided by the Lean 4 verifier. Crucially, verifier feedback, such as indicating success or detailing specific errors, allows the LLM to become ``self-aware'' of the correctness of its own reasoning process and learn to reflexively correct errors. Leanabell-Prover-V2 directly optimizes LLM reasoning trajectories with multi-turn verifier interactions, together with feedback token masking for stable RL training and a simple reward strategy. Experiments show that Leanabell-Prover-V2 improves performance by 3.2% (pass@128) with Kimina-Prover-Preview-Distill-7B and 2.0% (pass@128) with DeepSeek-Prover-V2-7B on the MiniF2F test set. The source codes, curated data and models are available at: https://github.com/Leanabell-LM/Leanabell-Prover-V2.

Lei Yang, Wei Bi, Chenxi Sun et al.

On-policy reinforcement learning (RL) methods widely used for language model post-training, like Group Relative Policy Optimization (GRPO), often suffer from limited exploration and early saturation due to low sampling diversity. While off-policy data can help, current approaches that mix entire trajectories cause significant policy mismatch and instability. In this work, we propose the $\textbf{S}$ingle-sample Mix-p$\textbf{O}$licy $\textbf{U}$nified $\textbf{P}$aradigm (SOUP), a framework that unifies off- and on-policy learning within individual samples at the token level. It confines off-policy influence to the prefix of a generated sequence sampled from historical policies, while the continuation is generated on-policy. Through token-level importance ratios, SOUP effectively leverages off-policy information while preserving training stability. Extensive experiments demonstrate that SOUP consistently outperforms standard on-policy training and existing off-policy extensions. Our further analysis clarifies how our fine-grained, single-sample mix-policy training can improve both exploration and final performance in LLM RL.

Jingyuan Zhang, Hongzhi Zhang, Zhou Haonan et al.

Data curation plays a crucial role in training powerful Visual Language Models (VLMs). In this work, we introduce the concept of Data Metabolism and present our data-centric framework to build VLMs throughout the development lifecycle. Starting from a standard model architecture, we discuss and provide insights into two crucial development steps: data curation and iteration, forming a closed-loop system that continuously improves model performance. We show a detailed codebook on how to process existing massive datasets and build user-specific data flywheel. As a demonstration, we release a VLM, named Capybara-VL, which excels in typical multimodal tasks (e.g. , visual question answering, scientific reasoning, and text-rich tasks). Despite its relatively compact size, Capybara-VL surpasses several open-source models that are up to 10 times larger in size. Moreover, it achieves results that are on par with those of several leading proprietary models, demonstrating its remarkable competitiveness. These results highlight the power of our data-centric framework and the potential of training smaller and more efficient VLMs.

Linhao Yu, Xinguang Ji, Yahui Liu et al.

Video captioning can be used to assess the video understanding capabilities of Multimodal Large Language Models (MLLMs). However, existing benchmarks and evaluation protocols suffer from crucial issues, such as inadequate or homogeneous creation of key points, exorbitant cost of data creation, and limited evaluation scopes. To address these issues, we propose an automatic framework, named AutoCaption, which leverages Monte Carlo Tree Search (MCTS) to construct numerous and diverse descriptive sentences (\textit{i.e.}, key points) that thoroughly represent video content in an iterative way. This iterative captioning strategy enables the continuous enhancement of video details such as actions, objects' attributes, environment details, etc. We apply AutoCaption to curate MCTS-VCB, a fine-grained video caption benchmark covering video details, thereby enabling a comprehensive evaluation of MLLMs on the video captioning task. We evaluate more than 20 open- and closed-source MLLMs of varying sizes on MCTS-VCB. Results show that MCTS-VCB can effectively and comprehensively evaluate the video captioning capability, with Gemini-1.5-Pro achieving the highest F1 score of 71.2. Interestingly, we fine-tune InternVL2.5-8B with the AutoCaption-generated data, which helps the model achieve an overall improvement of 25.0% on MCTS-VCB and 16.3% on DREAM-1K, further demonstrating the effectiveness of AutoCaption. The code and data are available at https://github.com/tjunlp-lab/MCTS-VCB.

Xingguang Ji, Jiakang Wang, Hongzhi Zhang et al.

With the development of Multimodal Large Language Models (MLLMs), numerous outstanding accomplishments have emerged within the open-source community. Due to the complexity of creating and training multimodal data pairs, it is still a computational and time-consuming process to build powerful MLLMs. In this work, we introduce Capybara-OMNI, an MLLM that trains in a lightweight and efficient manner and supports understanding text, image, video, and audio modalities. We present in detail the framework design, the data construction, and the training recipe, to develop an MLLM step-by-step to obtain competitive performance. We also provide exclusive benchmarks utilized in our experiments to show how to properly verify understanding capabilities across different modalities. Results show that by following our guidance, we can efficiently build an MLLM that achieves competitive performance among models of the same scale on various multimodal benchmarks. Additionally, to enhance the multimodal instruction following and conversational capabilities of the model, we further discuss how to train the chat version upon an MLLM understanding model, which is more in line with user habits for tasks like real-time interaction with humans. We publicly disclose the Capybara-OMNI model, along with its chat-based version. The disclosure includes both the model weights, a portion of the training data, and the inference codes, which are made available on GitHub.

Kangning Shen, Jingyuan Zhang, Chenxi Sun et al.

Large Language Models (LLMs) have demonstrated significant potential as autonomous software engineering (SWE) agents. Recent work has further explored augmenting these agents with memory mechanisms to support long-horizon reasoning. However, these approaches typically operate at a coarse instance granularity, treating the entire problem-solving episode as the atomic unit of storage and retrieval. We empirically demonstrate that instance-level memory suffers from a fundamental granularity mismatch, resulting in misguided retrieval when tasks with similar surface descriptions require distinct reasoning logic at specific stages. To address this, we propose Structurally Aligned Subtask-Level Memory, a method that aligns memory storage, retrieval, and updating with the agent's functional decomposition. Extensive experiments on SWE-bench Verified demonstrate that our method consistently outperforms both vanilla agents and strong instance-level memory baselines across diverse backbones, improving mean Pass@1 over the vanilla agent by +4.7 pp on average (e.g., +6.8 pp on Gemini 2.5 Pro). Performance gains grow with more interaction steps, showing that leveraging past experience benefits long-horizon reasoning in complex software engineering tasks.

Tongxin Li, Chenxi Sun

We tackle the challenge of learning to charge Electric Vehicles (EVs) with Out-of-Distribution (OOD) data. Traditional scheduling algorithms typically fail to balance near-optimal average performance with worst-case guarantees, particularly with OOD data. Model Predictive Control (MPC) is often too conservative and data-independent, whereas Reinforcement Learning (RL) tends to be overly aggressive and fully trusts the data, hindering their ability to consistently achieve the best-of-both-worlds. To bridge this gap, we introduce a novel OOD-aware scheduling algorithm, denoted OOD-Charging. This algorithm employs a dynamic "awareness radius", which updates in real-time based on the Temporal Difference (TD)-error that reflects the severity of OOD. The OOD-Charging algorithm allows for a more effective balance between consistency and robustness in EV charging schedules, thereby significantly enhancing adaptability and efficiency in real-world charging environments. Our results demonstrate that this approach improves the scheduling reward reliably under real OOD scenarios with remarkable shifts of EV charging behaviors caused by COVID-19 in the Caltech ACN-Data.

Chenxi Sun, Hongzhi Zhang, Qi Wang et al.

Instruction-based Image Editing (IIE) models have made significantly improvement due to the progress of multimodal large language models (MLLMs) and diffusion models, which can understand and reason about complex editing instructions. In addition to advancing current IIE models, accurately evaluating their output has become increasingly critical and challenging. Current IIE evaluation methods and their evaluation procedures often fall short of aligning with human judgment and often lack explainability. To address these limitations, we propose JUdgement through Routing of Expertise (JURE). Each expert in JURE is a pre-selected model assumed to be equipped with an atomic expertise that can provide useful feedback to judge output, and the router dynamically routes the evaluation task of a given instruction and its output to appropriate experts, aggregating their feedback into a final judge. JURE is trustworthy in two aspects. First, it can effortlessly provide explanations about its judge by examining the routed experts and their feedback. Second, experimental results demonstrate that JURE is reliable by achieving superior alignment with human judgments, setting a new standard for automated IIE evaluation. Moreover, JURE's flexible design is future-proof - modular experts can be seamlessly replaced or expanded to accommodate advancements in IIE, maintaining consistently high evaluation quality. Our evaluation data and results are available at https://github.com/Cyyyyyrus/JURE.git.

Chenxi Sun, Hongyan Li, Moxian Song et al.

Time series widely exists in real-world applications and many deep learning models have performed well on it. Current research has shown the importance of learning strategy for models, suggesting that the benefit is the order and size of learning samples. However, no effective strategy has been proposed for time series due to its abstract and dynamic construction. Meanwhile, the existing one-shot tasks and continuous tasks for time series necessitate distinct learning processes and mechanisms. No all-purpose approach has been suggested. In this work, we propose a novel Curricular and CyclicaL loss (CRUCIAL) to learn time series for the first time. It is model- and task-agnostic and can be plugged on top of the original loss with no extra procedure. CRUCIAL has two characteristics: It can arrange an easy-to-hard learning order by dynamically determining the sample contribution and modulating the loss amplitude; It can manage a cyclically changed dataset and achieve an adaptive cycle by correlating the loss distribution and the selection probability. We prove that compared with monotonous size, cyclical size can reduce expected error. Experiments on 3 kinds of tasks and 5 real-world datasets show the benefits of CRUCIAL for most deep learning models when learning time series.

Zeyu Xia, Chenxi Sun, Tianyu Xin et al.

Pansharpening aims to fuse high-resolution panchromatic (PAN) images with low-resolution multispectral (LRMS) images to generate high-resolution multispectral (HRMS) images. Although deep learning-based methods have achieved promising performance, they generally suffer from severe performance degradation when applied to data from unseen sensors. Adapting these models through full-scale retraining or designing more complex architectures is often prohibitively expensive and impractical for real-world deployment. To address this critical challenge, we propose a fast and general-purpose framework for cross-sensor adaptation, SWIFT (Sensitive Weight Identification for Fast Transfer). Specifically, SWIFT employs an unsupervised sampling strategy based on data manifold structures to balance sample selection while mitigating the bias of traditional Farthest Point Sampling, efficiently selecting only 3\% of the most informative samples from the target domain. This subset is then used to probe a source-domain pre-trained model by analyzing the gradient behavior of its parameters, allowing for the quick identification and subsequent update of only the weight subset most sensitive to the domain shift. As a plug-and-play framework, SWIFT can be applied to various existing pansharpening models. Extensive experiments demonstrate that SWIFT reduces the adaptation time from hours to approximately one minute on a single NVIDIA RTX 4090 GPU. The adapted models not only substantially outperform direct-transfer baselines but also achieve performance competitive with, and in some cases superior to, full retraining, establishing a new state-of-the-art on cross-sensor pansharpening tasks for the WorldView-2 and QuickBird datasets.

Ying Zhang, Shuai Guo, Chenxi Sun et al.

In recent years, deformable medical image registration techniques have made significant progress. However, existing models still lack efficiency in parallel extraction of coarse and fine-grained features. To address this, we construct a new pyramid registration network based on feature and deformation field (FF-PNet). For coarse-grained feature extraction, we design a Residual Feature Fusion Module (RFFM), for fine-grained image deformation, we propose a Residual Deformation Field Fusion Module (RDFFM). Through the parallel operation of these two modules, the model can effectively handle complex image deformations. It is worth emphasizing that the encoding stage of FF-PNet only employs traditional convolutional neural networks without any attention mechanisms or multilayer perceptrons, yet it still achieves remarkable improvements in registration accuracy, fully demonstrating the superior feature decoding capabilities of RFFM and RDFFM. We conducted extensive experiments on the LPBA and OASIS datasets. The results show our network consistently outperforms popular methods in metrics like the Dice Similarity Coefficient.

Chenxi Sun, Hongyan Li, Yaliang Li et al.

Data is essential to performing time series analysis utilizing machine learning approaches, whether for classic models or today's large language models. A good time-series dataset is advantageous for the model's accuracy, robustness, and convergence, as well as task outcomes and costs. The emergence of data-centric AI represents a shift in the landscape from model refinement to prioritizing data quality. Even though time-series data processing methods frequently come up in a wide range of research fields, it hasn't been well investigated as a specific topic. To fill the gap, in this paper, we systematically review different data-centric methods in time series analysis, covering a wide range of research topics. Based on the time-series data characteristics at sample, feature, and period, we propose a taxonomy for the reviewed data selection methods. In addition to discussing and summarizing their characteristics, benefits, and drawbacks targeting time-series data, we also introduce the challenges and opportunities by proposing recommendations, open problems, and possible research topics.

Pengju Zhang, Chenxi Sun, Jianwei Zhang et al.

We have developed an individual identification system based on magnetocardiography (MCG) signals captured using optically pumped magnetometers (OPMs). Our system utilizes pattern recognition to analyze the signals obtained at different positions on the body, by scanning the matrices composed of MCG signals with a 2*2 window. In order to make use of the spatial information of MCG signals, we transform the signals from adjacent small areas into four channels of a dataset. We further transform the data into time-frequency matrices using wavelet transforms and employ a convolutional neural network (CNN) for classification. As a result, our system achieves an accuracy rate of 97.04% in identifying individuals. This finding indicates that the MCG signal holds potential for use in individual identification systems, offering a valuable tool for personalized healthcare management.

Chenxi Sun, Hongyan Li, Moxian Song et al.

Spiking Neural Networks (SNNs) have a greater potential for modeling time series data than Artificial Neural Networks (ANNs), due to their inherent neuron dynamics and low energy consumption. However, it is difficult to demonstrate their superiority in classification accuracy, because current efforts mainly focus on designing better network structures. In this work, enlighten by brain-inspired science, we find that, not only the structure but also the learning process should be human-like. To achieve this, we investigate the power of Curriculum Learning (CL) on SNNs by designing a novel method named CSNN with two theoretically guaranteed mechanisms: The active-to-dormant training order makes the curriculum similar to that of human learning and suitable for spiking neurons; The value-based regional encoding makes the neuron activity to mimic the brain memory when learning sequential data. Experiments on multiple time series sources including simulated, sensor, motion, and healthcare demonstrate that CL has a more positive effect on SNNs than ANNs with about twice the accuracy change, and CSNN can increase about 3% SNNs' accuracy by improving network sparsity, neuron firing status, anti-noise ability, and convergence speed.

Yi Zhong, Chen Zhang, Xule Liu et al.

While Current TTS systems perform well in synthesizing high-quality speech, producing highly expressive speech remains a challenge. Emphasis, as a critical factor in determining the expressiveness of speech, has attracted more attention nowadays. Previous works usually enhance the emphasis by adding intermediate features, but they can not guarantee the overall expressiveness of the speech. To resolve this matter, we propose Emphatic Expressive TTS (EE-TTS), which leverages multi-level linguistic information from syntax and semantics. EE-TTS contains an emphasis predictor that can identify appropriate emphasis positions from text and a conditioned acoustic model to synthesize expressive speech with emphasis and linguistic information. Experimental results indicate that EE-TTS outperforms baseline with MOS improvements of 0.49 and 0.67 in expressiveness and naturalness. EE-TTS also shows strong generalization across different datasets according to AB test results.

Yen-Hsiu Chou, Shenda Hong, Chenxi Sun et al.

Since data is presented long-tailed in reality, it is challenging for Federated Learning (FL) to train across decentralized clients as practical applications. We present Global-Regularized Personalization (GRP-FED) to tackle the data imbalanced issue by considering a single global model and multiple local models for each client. With adaptive aggregation, the global model treats multiple clients fairly and mitigates the global long-tailed issue. Each local model is learned from the local data and aligns with its distribution for customization. To prevent the local model from just overfitting, GRP-FED applies an adversarial discriminator to regularize between the learned global-local features. Extensive results show that our GRP-FED improves under both global and local scenarios on real-world MIT-BIH and synthesis CIFAR-10 datasets, achieving comparable performance and addressing client imbalance.

Chenxi Sun, Shenda Hong, Moxian Song et al.

Prediction based on Irregularly Sampled Time Series (ISTS) is of wide concern in the real-world applications. For more accurate prediction, the methods had better grasp more data characteristics. Different from ordinary time series, ISTS is characterised with irregular time intervals of intra-series and different sampling rates of inter-series. However, existing methods have suboptimal predictions due to artificially introducing new dependencies in a time series and biasedly learning relations among time series when modeling these two characteristics. In this work, we propose a novel Time Encoding (TE) mechanism. TE can embed the time information as time vectors in the complex domain. It has the the properties of absolute distance and relative distance under different sampling rates, which helps to represent both two irregularities of ISTS. Meanwhile, we create a new model structure named Time Encoding Echo State Network (TE-ESN). It is the first ESNs-based model that can process ISTS data. Besides, TE-ESN can incorporate long short-term memories and series fusion to grasp horizontal and vertical relations. Experiments on one chaos system and three real-world datasets show that TE-ESN performs better than all baselines and has better reservoir property.

Chenxi Sun, Moxian Song, Shenda Hong et al.

Recurrent Neural Networks (RNNs) have demonstrated their outstanding ability in sequence tasks and have achieved state-of-the-art in wide range of applications, such as industrial, medical, economic and linguistic. Echo State Network (ESN) is simple type of RNNs and has emerged in the last decade as an alternative to gradient descent training based RNNs. ESN, with a strong theoretical ground, is practical, conceptually simple, easy to implement. It avoids non-converging and computationally expensive in the gradient descent methods. Since ESN was put forward in 2002, abundant existing works have promoted the progress of ESN, and the recently introduced Deep ESN model opened the way to uniting the merits of deep learning and ESNs. Besides, the combinations of ESNs with other machine learning models have also overperformed baselines in some applications. However, the apparent simplicity of ESNs can sometimes be deceptive and successfully applying ESNs needs some experience. Thus, in this paper, we categorize the ESN-based methods to basic ESNs, DeepESNs and combinations, then analyze them from the perspective of theoretical studies, network designs and specific applications. Finally, we discuss the challenges and opportunities of ESNs by summarizing the open questions and proposing possible future works.

Chenxi Sun, Shenda Hong, Moxian Song et al.

Irregularly sampled time series (ISTS) data has irregular temporal intervals between observations and different sampling rates between sequences. ISTS commonly appears in healthcare, economics, and geoscience. Especially in the medical environment, the widely used Electronic Health Records (EHRs) have abundant typical irregularly sampled medical time series (ISMTS) data. Developing deep learning methods on EHRs data is critical for personalized treatment, precise diagnosis and medical management. However, it is challenging to directly use deep learning models for ISMTS data. On the one hand, ISMTS data has the intra-series and inter-series relations. Both the local and global structures should be considered. On the other hand, methods should consider the trade-off between task accuracy and model complexity and remain generality and interpretability. So far, many existing works have tried to solve the above problems and have achieved good results. In this paper, we review these deep learning methods from the perspectives of technology and task. Under the technology-driven perspective, we summarize them into two categories - missing data-based methods and raw data-based methods. Under the task-driven perspective, we also summarize them into two categories - data imputation-oriented and downstream task-oriented. For each of them, we point out their advantages and disadvantages. Moreover, we implement some representative methods and compare them on four medical datasets with two tasks. Finally, we discuss the challenges and opportunities in this area.