The control of nonlinear dynamical systems remains a major challenge for autonomous agents. Current trends in reinforcement learning (RL) focus on complex representations of dynamics and policies, which have yielded impressive results in solving a variety of hard control tasks. However, this new sophistication and extremely over-parameterized models have come with the cost of an overall reduction in our ability to interpret the resulting policies. In this paper, we take inspiration from the control community and apply the principles of hybrid switching systems in order to break down complex dynamics into simpler components. We exploit the rich representational power of probabilistic graphical models and derive an expectation-maximization (EM) algorithm for learning a sequence model to capture the temporal structure of the data and automatically decompose nonlinear dynamics into stochastic switching linear dynamical systems. Moreover, we show how this framework of switching models enables extracting hierarchies of Markovian and auto-regressive locally linear controllers from nonlinear experts in an imitation learning scenario.
Identification of Nonlinear Latent Hierarchical Models
Carnegie Mellon University（卡内基梅隆大学）；University of California San Diego（加州大学圣迭戈分校）；Beijing Technology and Business University（北京工商大学）；Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）
The focus in this paper is Bayesian system identification based on noisy incomplete modal data where we can impose spatially-sparse stiffness changes when updating a structural model. To this end, based on a similar hierarchical sparse Bayesian learning model from our previous work, we propose two Gibbs sampling algorithms. The algorithms differ in their strategies to deal with the posterior uncertainty of the equation-error precision parameter, but both sample from the conditional posterior probability density functions (PDFs) for the structural stiffness parameters and system modal parameter…
We propose a new statistical protocol for the estimation of precipitation using lightning data. We first identify rainy events using a scan statistics, then we estimate Rainfall Lighting Ratio (RLR) to convert lightning number into rain volume given the storm intensity. Then we build a hierarchical Bayesian model aiming at the prediction of 15- and 30-minutes cumulated precipitation at unobserved locations and time using information on lightning in the same area. More specifically, we build a Bayesian hierarchical model in which precipitation is modeled as function of lightning count and space…
The light we receive from distant astrophysical objects carries information about their origins and the physical mechanisms that power them. The study of these signals, however, is complicated by the fact that observations are often a mixture of the light emitted by multiple localized sources situated in a spatially-varying background. A general algorithm to achieve robust and accurate source identification in this case remains an open question in astrophysics. This paper focuses on high-energy light (such as X-rays and gamma-rays), for which observatories can detect individual photons (quanta…
The remarkable performance of the pre-trained language model (LM) using self-supervised learning has led to a major paradigm shift in the study of natural language processing. In line with these changes, leveraging the performance of speech recognition systems with massive deep learning-based LMs is a major topic of speech recognition research. Among the various methods of applying LMs to speech recognition systems, in this paper, we focus on a cross-modal knowledge distillation method that transfers knowledge between two types of deep neural networks with different modalities. We propose an a…
Sun Yat-sen University（中山大学）；Alibaba Group（阿里巴巴集团）
On many natural language processing tasks, large pre-trained language models (PLMs) have shown overwhelming performances compared with traditional neural network methods. Nevertheless, their huge model size and low inference speed have hindered the deployment on resource-limited devices in practice. In this paper, we target to compress PLMs with knowledge distillation, and propose a hierarchical relational knowledge distillation (HRKD) method to capture both hierarchical and domain relational information. Specifically, to enhance the model capability and transferability, we leverage the idea o…
NTT Media Intelligence Laboratories, NTT Corporation（NTT公司NTT媒体情报实验室）
This paper presents a novel knowledge distillation method for dialogue sequence labeling. Dialogue sequence labeling is a supervised learning task that estimates labels for each utterance in the target dialogue document, and is useful for many applications such as dialogue act estimation. Accurate labeling is often realized by a hierarchically-structured large model consisting of utterance-level and dialogue-level networks that capture the contexts within an utterance and between utterances, respectively. However, due to its large model size, such a model cannot be deployed on resource-constra…
Federated learning (FL) has been recognized as a privacy-preserving distributed machine learning paradigm that enables knowledge sharing among various heterogeneous artificial intelligence (AIoT) devices through centralized global model aggregation. FL suffers from model inaccuracy and slow convergence due to the model heterogeneity of the AIoT devices involved. Although various existing methods try to solve the bottleneck of the model heterogeneity problem, most of them improve the accuracy of heterogeneous models in a coarse-grained manner, which makes it still a great challenge to deploy la…
NTT Media Intelligence Laboratories, NTT Corporation（NTT公司NTT媒体情报实验室）
We present a novel large-context end-to-end automatic speech recognition (E2E-ASR) model and its effective training method based on knowledge distillation. Common E2E-ASR models have mainly focused on utterance-level processing in which each utterance is independently transcribed. On the other hand, large-context E2E-ASR models, which take into account long-range sequential contexts beyond utterance boundaries, well handle a sequence of utterances such as discourses and conversations. However, the transformer architecture, which has recently achieved state-of-the-art ASR performance among utte…
Music Technology Group, Universitat Pompeu Fabra（庞培法布拉大学音乐技术组）；Dolby Laboratories（杜比实验室）；European Commission, Joint Research Centre（欧盟委员会联合研究中心）
Version identification systems aim to detect different renditions of the same underlying musical composition (loosely called cover songs). By learning to encode entire recordings into plain vector embeddings, recent systems have made significant progress in bridging the gap between accuracy and scalability, which has been a key challenge for nearly two decades. In this work, we propose to further narrow this gap by employing a set of data distillation techniques that reduce the embedding dimensionality of a pre-trained state-of-the-art model. We compare a wide range of techniques and propose n…
Person re-identification (ReID) is aimed at identifying the same person across videos captured from different cameras. In the view that networks extracting global features using ordinary network architectures are difficult to extract local features due to their weak attention mechanisms, researchers have proposed a lot of elaborately designed ReID networks, while greatly improving the accuracy, the model size and the feature extraction latency are also soaring. We argue that a relatively compact ordinary network extracting globally pooled features has the capability to extract discriminative l…
University of Modena and Reggio Emilia（摩德纳和雷焦艾米利亚大学）
To achieve robustness in Re-Identification, standard methods leverage tracking information in a Video-To-Video fashion. However, these solutions face a large drop in performance for single image queries (e.g., Image-To-Video setting). Recent works address this severe degradation by transferring temporal information from a Video-based network to an Image-based one. In this work, we devise a training strategy that allows the transfer of a superior knowledge, arising from a set of views depicting the target object. Our proposal - Views Knowledge Distillation (VKD) - pins this visual variety as a …
Deep Medicine, University of Oxford（牛津大学Deep Medicine（深度医学））；University of Oxford（牛津大学）
Recent evidence shows that deep learning models trained on electronic health records from millions of patients can deliver substantially more accurate predictions of risk compared to their statistical counterparts. While this provides an important opportunity for improving clinical decision-making, the lack of interpretability is a major barrier to the incorporation of these black-box models in routine care, limiting their trustworthiness and preventing further hypothesis-testing investigations. In this study, we propose two methods, namely, model distillation and variable selection, to untang…
Predictive Free Energy Simulations Through Hierarchical Distillation of Quantum Hamiltonians
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA, 91125, USA（化学与化工系，加州理工学院，帕萨迪纳，CA，91125，USA）；Chemical Engineering, California Institute of Technology, Pasadena, CA, 91125, USA（化学工程，加州理工学院，帕萨迪纳，CA，91125，USA）
Hierarchical Multi-to-Single-Modal Knowledge Distillation for Disruption Prediction in EAST
School of Computer Science and Technology, Anhui University（安徽大学计算机科学与技术学院）；Institute of Plasma Physics, Chinese Academy of Sciences（中国科学院等离子体物理研究所）；State Key Laboratory of Opto-Electronic Information Acquisition and Protection Technology, Anhui University（安徽大学光电子信息获取与控制技术国家重点实验室）
We propose a lifelong learning system that has the ability to reuse and transfer knowledge from one task to another while efficiently retaining the previously learned knowledge-base. Knowledge is transferred by learning reusable skills to solve tasks in Minecraft, a popular video game which is an unsolved and high-dimensional lifelong learning problem. These reusable skills, which we refer to as Deep Skill Networks, are then incorporated into our novel Hierarchical Deep Reinforcement Learning Network (H-DRLN) architecture using two techniques: (1) a deep skill array and (2) skill distillation,…
Zhejiang University（浙江大学）；China Electric Power Research Institute（中国电力科学研究院）
Insights in power grid pixel maps (PGPMs) refer to important facility operating states and unexpected changes in the power grid. Identifying insights helps analysts understand the collaboration of various parts of the grid so that preventive and correct operations can be taken to avoid potential accidents. Existing solutions for identifying insights in PGPMs are performed manually, which may be laborious and expertise-dependent. In this paper, we propose an interactive insight identification and annotation framework by leveraging an enhanced variational autoencoder (VAE). In particular, a new …
Instituto Federal do Paraná (IFPR)（巴拉那联邦学院）；Universidade Estadual de Maringá (UEM)（马林加州立大学）
The COVID-19 can cause severe pneumonia and is estimated to have a high impact on the healthcare system. The standard image diagnosis tests for pneumonia are chest X-ray (CXR) and computed tomography (CT) scan. CXR are useful in because it is cheaper, faster and more widespread than CT. This study aims to identify pneumonia caused by COVID-19 from other types and also healthy lungs using only CXR images. In order to achieve the objectives, we have proposed a classification schema considering the multi-class and hierarchical perspectives, since pneumonia can be structured as a hierarchy. Given …
McGill University（麦吉尔大学）；Centre for Intelligent Machines (CIM)（智能机器中心 (CIM)）；Department of Electrical and Computer Engineering（电气与计算机工程系）
Segmentation of enhancing tumours or lesions from MRI is important for detecting new disease activity in many clinical contexts. However, accurate segmentation requires the inclusion of medical images (e.g., T1 post contrast MRI) acquired after injecting patients with a contrast agent (e.g., Gadolinium), a process no longer thought to be safe. Although a number of modality-agnostic segmentation networks have been developed over the past few years, they have been met with limited success in the context of enhancing pathology segmentation. In this work, we present HAD-Net, a novel offline advers…
Acoustic identification of individual animals with hierarchical contrastive learning
Centre for Digital Music, Queen Mary University of London（伦敦大学玛丽皇后学院数字音乐中心）；IMT Atlantique（IMT大西洋高等电信学院）；CNRS（法国国家科学研究中心）；Lab-STICC（信息、信号与智能系统实验室）；Tilburg University（蒂尔堡大学）
SenseTime Research（商汤科技研究院）；National University of Singapore（新加坡国立大学）
Person Re-Identification (Re-ID) is of great importance to the many video surveillance systems. Learning discriminative features for Re-ID remains a challenge due to the large variations in the image space, e.g., continuously changing human poses, illuminations and point of views. In this paper, we propose HAVANA, a novel extensible, light-weight HierArchical and VAriation-Normalized Autoencoder that learns features robust to intra-class variations. In contrast to existing generative approaches that prune the variations with heavy extra supervised signals, HAVANA suppresses the intra-class var…
Recent works in the person re-identification task mainly focus on the model accuracy while ignore factors related to the efficiency, e.g. model size and latency, which are critical for practical application. In this paper, we propose a novel Hierarchical and Efficient Network (HENet) that learns hierarchical global, partial, and recovery features ensemble under the supervision of multiple loss combinations. To further improve the robustness against the irregular occlusion, we propose a new dataset augmentation approach, dubbed Random Polygon Erasing (RPE), to random erase irregular area of the…
Target detection in hyperspectral imagery is the process of locating pixels from an image which are likely to contain target, typically done by comparing one or more spectra for the desired target material to each pixel in the image. Target identification is the process of target detection incorporating an additional process to identify more specifically the material that is present in each pixel that scored high in detection. Detection is generally a 2-class problem of target vs. background, and identification is a many class problem including target, background, and additional know materials…
Sun Yat-Sen University（中山大学）；University of Alberta（阿尔伯塔大学）
Knowledge distillation is an effective technique that transfers knowledge from a large teacher model to a shallow student. However, just like massive classification, large scale knowledge distillation also imposes heavy computational costs on training models of deep neural networks, as the softmax activations at the last layer involve computing probabilities over numerous classes. In this work, we apply the idea of importance sampling which is often used in Neural Machine Translation on large scale knowledge distillation. We present a method called dynamic importance sampling, where ranked cla…
Pohang University of Science and Technology (POSTECH)（浦项科技大学）
Recommender Systems (RS) have employed knowledge distillation which is a model compression technique training a compact student model with the knowledge transferred from a pre-trained large teacher model. Recent work has shown that transferring knowledge from the teacher's intermediate layer significantly improves the recommendation quality of the student. However, they transfer the knowledge of individual representation point-wise and thus have a limitation in that primary information of RS lies in the relations in the representation space. This paper proposes a new topology distillation appr…
Wuhan University of Technology（武汉理工大学）；University of Surrey（萨里大学）；Wuhan University（武汉大学）
Motivation: Identifying interaction clusters of large gene regulatory networks (GRNs) is critical for its further investigation, while this task is very challenging, attributed to data noise in experiment data, large scale of GRNs, and inconsistency between gene expression profiles and function modules, etc. It is promising to semi-supervise this process by prior information, but shortage of prior information sometimes make it very challenging. Meanwhile, it is also annoying, and sometimes impossible to discovery gold standard for evaluation of clustering results.\\ Results: With assistance of…
Pune Institute of Computer Technology（浦那计算机技术学院）；Indian Institute of Technology Madras（印度理工学院马德拉斯分校）
With the open-source revolution, source codes are now more easily accessible than ever. This has, however, made it easier for malicious users and institutions to copy the code without giving regards to the license, or credit to the original author. Therefore, source code author identification is a critical task with paramount importance. In this paper, we propose ICodeNet - a hierarchical neural network that can be used for source code file-level tasks. The ICodeNet processes source code in image format and is employed for the task of per file author identification. The ICodeNet consists of an…
Leipzig University of Applied Sciences (HTWK)（莱比锡应用科学大学）
This paper introduces a new unsupervised method for the clustering of physiological data into health states based on their similarity. We propose an iterative hierarchical clustering approach that combines health states according to a similarity constraint to new arbitrary health states. We applied method to experimental data in which the physical strain of subjects was systematically varied. We derived health states based on parameters extracted from ECG data. The occurrence of health states shows a high temporal correlation to the experimental phases of the physical exercise. We compared our…
University of Eastern Finland（东芬兰大学）；AIST（日本产业技术综合研究所）
Language recognition system is typically trained directly to optimize classification error on the target language labels, without using the external, or meta-information in the estimation of the model parameters. However labels are not independent of each other, there is a dependency enforced by, for example, the language family, which affects negatively on classification. The other external information sources (e.g. audio encoding, telephony or video speech) can also decrease classification accuracy. In this paper, we attempt to solve these issues by constructing a deep hierarchical neural ne…
Shared Attention-based Autoencoder with Hierarchical Fusion-based Graph Convolution Network for sEEG SOZ Identification
School of Electronic and Information Engineering, South China University of Technology（华南理工大学电子与信息工程学院）；Fujian Medical University Union Hospital（福建医科大学附属协和医院）
Hierarchical Multi-Graphs Learning for Robust Group Re-Identification
College of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；College of Big Data and Software, Chongqing University（重庆大学大数据与软件学院）；University of Michigan - Ann Arbor（密歇根大学安娜堡分校）
Bangladesh University of Engineering and Technology（孟加拉国工程技术大学）
Digital security has been an active area of research interest due to the rapid adaptation of internet infrastructure, the increasing popularity of social media, and digital cameras. Due to inherent differences in working principles to generate an image, different camera brands left behind different intrinsic processing noises which can be used to identify the camera brand. In the last decade, many signal processing and deep learning-based methods have been proposed to identify and isolate this noise from the scene details in an image to detect the source camera brand. One prominent solution is…
Identifying multiple speakers without knowing where a speaker's voice is in a recording is a challenging task. This paper proposes a hierarchical network with transformer encoders and memory mechanism to address this problem. The proposed model contains a frame-level encoder and segment-level encoder, both of them make use of the transformer encoder block. The multi-head attention mechanism in the transformer structure could better capture different speaker properties when the input utterance contains multiple speakers. The memory mechanism used in the frame-level encoders can build a recurren…
