# 机器学习学术速递[8.11]

Update！H5支撑摘要折叠，体会更佳！点击拜访arxivdaily.com cs.LG 方向，今天合计64篇 Graph相关(图学习|图神经网络|图优化等)(2篇) 【1】 L…

Update！H5支撑摘要折叠，体会更佳！点击拜访arxivdaily.com

cs.LG 方向，今天合计64篇

Graph相关(图学习|图神经网络|图优化等)(2篇)

【1】 Label-informed Graph Structure Learning for Node Classification

：Liping Wang,Fenyu Hu,Shu Wu,Liang Wang

domains. Nevertheless, most GNN methods are sensitive to the quality of graph
structures. To tackle this problem, some studies exploit different graph
structure learning strategies to refine the original graph structure. However,
these methods only consider feature information while ignoring available label
information. In this paper, we propose a novel label-informed graph structure
learning framework which incorporates label information explicitly through a
class transition matrix. We conduct extensive experiments on seven node
classification benchmark datasets and the results show that our method
outperforms or matches the state-of-the-art baselines.

【2】 Learning Multi-Granular Spatio-Temporal Graph Network for Skeleton-based  Action Recognition

：Tailin Chen,Desen Zhou,Jian Wang,Shidong Wang,Yu Guan,Xuming He,Errui Ding

GAN|对立|进犯|生成相关(9篇)

【1】 Continual Learning for Grounded Instruction Generation by Observing  Human Following Behavior

：Noriyuki Kojima,Alane Suhr,Yoav Artzi

observing human users' instruction execution. We focus on a collaborative
scenario, where the system both acts and delegates tasks to human users using
natural language. We compare user execution of generated instructions to the
original system intent as an indication to the system's success communicating
its intent. We show how to use this signal to improve the system's ability to
generate instructions via contextual bandit learning. In interaction with real
users, our system demonstrates dramatic improvements in its ability to generate
language over time.

【2】 Correlation Clustering Reconstruction in Semi-Adversarial Models

：Flavio Chierichetti,Alessandro Panconesi,Giuseppe Re,Luca Trevisan

applications. We study the reconstruction version of this problem in which one
is seeking to reconstruct a latent clustering that has been corrupted by random
noise and adversarial modifications.
Concerning the latter, we study a standard "post-adversarial" model, in which
adversarial modifications come after the noise, and also introduce and analyze
a "pre-adversarial" model in which adversarial modifications come before the
noise. Given an input coming from such a semi-adversarial generative model, the
goal is to reconstruct almost perfectly and with high probability the latent
clustering.
We focus on the case where the hidden clusters have equal size and show the
following. In the pre-adversarial setting, spectral algorithms are optimal, in
the sense that they reconstruct all the way to the information-theoretic
threshold beyond which no reconstruction is possible. In contrast, in the
post-adversarial setting their ability to restore the hidden clusters stops
before the threshold, but the gap is optimally filled by SDP-based algorithms.

【3】 UniNet: A Unified Scene Understanding Network and Exploring Multi-Task  Relationships through the Lens of Adversarial Attacks

：NareshKumar Gurulingan,Elahe Arani,Bahram Zonooz

in the real world. Single task vision networks extract information only based
on some aspects of the scene. In multi-task learning (MTL), on the other hand,
these single tasks are jointly learned, thereby providing an opportunity for
tasks to share information and obtain a more comprehensive understanding. To
this end, we develop UniNet, a unified scene understanding network that
accurately and efficiently infers vital vision tasks including object
detection, semantic segmentation, instance segmentation, monocular depth
estimation, and monocular instance depth prediction. As these tasks look at
different semantic and geometric information, they can either complement or
conflict with each other. Therefore, understanding inter-task relationships can
provide useful cues to enable complementary information sharing. We evaluate
the task relationships in UniNet through the lens of adversarial attacks based
on the notion that they can exploit learned biases and task interactions in the
neural network. Extensive experiments on the Cityscapes dataset, using
untargeted and targeted attacks reveal that semantic tasks strongly interact
amongst themselves, and the same holds for geometric tasks. Additionally, we
show that the relationship between semantic and geometric tasks is asymmetric
and their interaction becomes weaker as we move towards higher-level
representations.

【4】 Regularized Sequential Latent Variable Models with Adversarial Neural  Networks

：Jin Huang,Ming Xiao

【5】 Enhancing Knowledge Tracing via Adversarial Training

：Xiaopeng Guo,Zhijie Huang,Jie Gao,Mingyu Shang,Maojing Shu,Jun Sun

students' knowledge mastery over time so as to make predictions on their future
performance. Owing to the good representation capacity of deep neural networks
(DNNs), recent advances on KT have increasingly concentrated on exploring DNNs
to improve the performance of KT. However, we empirically reveal that the DNNs
based KT models may run the risk of overfitting, especially on small datasets,
leading to limited generalization. In this paper, by leveraging the current
advances in adversarial training (AT), we propose an efficient AT based KT
method (ATKT) to enhance KT model's generalization and thus push the limit of
KT. Specifically, we first construct adversarial perturbations and add them on
the original interaction embeddings as adversarial examples. The original and
adversarial examples are further used to jointly train the KT model, forcing it
is not only to be robust to the adversarial examples, but also to enhance the
generalization over the original ones. To better implement AT, we then present
an efficient attentive-LSTM model as KT backbone, where the key is a proposed
knowledge hidden state attention module that adaptively aggregates information
from previous knowledge hidden states while simultaneously highlighting the
importance of current knowledge hidden state to make a more accurate
prediction. Extensive experiments on four public benchmark datasets demonstrate
that our ATKT achieves new state-of-the-art performance. Code is available at:
\color{blue} {\url{https://github.com/xiaopengguo/ATKT}}.

【6】 On Procedural Adversarial Noise Attack And Defense

：Jun Yan,Xiaoyang Deng,Huilin Yin,Wancheng Ge

would inveigle neural networks to make prediction errors with small per-
turbations on the input images. Researchers have been devoted to promoting the
research on the universal adversarial perturbations (UAPs) which are
gradient-free and have little prior knowledge on data distributions. Procedural
adversarial noise at- tack is a data-free universal perturbation generation
method. In this paper, we propose two universal adversarial perturbation (UAP)
generation methods based on procedural noise functions: Simplex noise and
Worley noise. In our framework, the shading which disturbs visual
classification is generated with rendering technology. Without changing the
semantic representations, the adversarial examples generated via our methods
show superior performance on the attack.

【7】 Adversarial Open Domain Adaption Framework (AODA): Sketch-to-Photo  Synthesis

：Amey Thakur,Mega Satish

【8】 Explainable AI and susceptibility to adversarial attacks: a case study  in classification of breast ultrasound images

：Hamza Rasaee,Hassan Rivaz

to classify suspicious breast nodules and potentially detect the onset of
breast cancer. Recently, Convolutional Neural Networks (CNN) techniques have
shown promising results in classifying ultrasound images of the breast into
benign or malignant. However, CNN inference acts as a black-box model, and as
such, its decision-making is not interpretable. Therefore, increasing effort
has been dedicated to explaining this process, most notably through GRAD-CAM
and other techniques that provide visual explanations into inner workings of
CNNs. In addition to interpretation, these methods provide clinically important
information, such as identifying the location for biopsy or treatment. In this
work, we analyze how adversarial assaults that are practically undetectable may
be devised to alter these importance maps dramatically. Furthermore, we will
show that this change in the importance maps can come with or without altering
the classification result, rendering them even harder to detect. As such, care
must be taken when using these importance maps to shed light on the inner
workings of deep learning. Finally, we utilize Multi-Task Learning (MTL) and
propose a new network based on ResNet-50 to improve the classification
accuracies. Our sensitivity and specificity is comparable to the state of the
art results.

【9】 TDLS: A Top-Down Layer Searching Algorithm for Generating Counterfactual  Visual Explanation

：Cong Wang,Haocheng Han,Caleb Chen Cao

【1】 R4Dyn: Exploring Radar for Self-Supervised Monocular Depth Estimation of  Dynamic Scenes

：Stefano Gasperini,Patrick Koch,Vinzenz Dallabetta,Nassir Navab,Benjamin Busam,Federico Tombari

【2】 Semi-supervised classification of radiology images with NoTeacher: A  Teacher that is not Mean

：Balagopal Unnikrishnan,Cuong Nguyen,Shafa Balaram,Chao Li,Chuan Sheng Foo,Pavitra Krishnaswamy

【3】 Active Learning for Transition State Calculation

：Shuting Gu,Hongqiao Wang,Xiang Zhou

【1】 AdaRNN: Adaptive Learning and Forecasting of Time Series

：Yuntao Du,Jindong Wang,Wenjie Feng,Sinno Pan,Tao Qin,Chongjun Wang

【2】 Deep Transfer Learning for Identifications of Slope Surface Cracks

：Yuting Yang,Gang Mei

【1】 A Survey on Deep Reinforcement Learning for Data Processing and  Analytics

：Qingpeng Cai,Can Cui,Yiyuan Xiong,Zhongle Xie,Meihui Zhang

【2】 Deep Reinforcement Learning for Demand Driven Services in Logistics and  Transportation Systems: A Survey

：Zefang Zong,Tao Feng,Tong Xia,Depeng,Yong Li

【3】 High Quality Related Search Query Suggestions using Deep Reinforcement  Learning

：Praveen Kumar Bodigutla

【4】 Adaptable image quality assessment using meta-reinforcement learning of  task amenability

：Shaheer U. Saeed,Yunguan Fu,Vasilis Stavrinides,Zachary M. C. Baum,Qianye Yang,Mirabela Rusu,Richard E. Fan,Geoffrey A. Sonn,J. Alison Noble,Dean C. Barratt,Yipeng Hu

with image data quality. When developing modern deep learning algorithms,
rather than relying on subjective (human-based) image quality assessment (IQA),
task amenability potentially provides an objective measure of task-specific
image quality. To predict task amenability, an IQA agent is trained using
reinforcement learning (RL) with a simultaneously optimised task predictor,
such as a classification or segmentation neural network. In this work, we
develop transfer learning or adaptation strategies to increase the adaptability
of both the IQA agent and the task predictor so that they are less dependent on
high-quality, expert-labelled training data. The proposed transfer learning
strategy re-formulates the original RL problem for task amenability in a
meta-reinforcement learning (meta-RL) framework. The resulting algorithm
facilitates efficient adaptation of the agent to different definitions of image
quality, each with its own Markov decision process environment including
different images, labels and an adaptable task predictor. Our work demonstrates
that the IQA agents pre-trained on non-expert task labels can be adapted to
predict task amenability as defined by expert task labels, using only a small
set of expert labels. Using 6644 clinical ultrasound images from 249 prostate
cancer patients, our results for image classification and segmentation tasks
show that the proposed IQA method can be adapted using data with as few as
respective 19.7% and 29.6% expert-reviewed consensus labels and still achieve
comparable IQA and task performance, which would otherwise require a training
dataset with 100% expert labels.

【5】 Imitation Learning by Reinforcement Learning

：Kamil Ciosek

behavior. Somewhat counterintuitively, we show that, for deterministic experts,
imitation learning can be done by reduction to reinforcement learning, which is
commonly considered more difficult. We conduct experiments which confirm that
our reduction works well in practice for a continuous control task.

【1】 Meta-repository of screening mammography classifiers

：Benjamin Stadnick,Jan Witowski,Vishwaesh Rajiv,Jakub Chłędowski,Farah E. Shamout,Kyunghyun Cho,Krzysztof J. Geras

【1】 Hierarchical Latent Relation Modeling for Collaborative Metric Learning

：Viet-Anh Tran,Guillaume Salha-Galvan,Romain Hennequin,Manuel Moussallam

【1】 Analyzing Effects of The COVID-19 Pandemic on Road Traffic Safety: The  Cases of New York City, Los Angeles, and Boston

：Lahari Karadla,Weizi Li

throughout the world. In addition to the health consequences, the impacts on
traffic behaviors have also been sudden and dramatic. We have analyzed how the
road traffic safety of New York City, Los Angeles, and Boston in the U.S. have
been impacted by the pandemic and corresponding local government orders and
restrictions. To be specific, we have studied the accident hotspots'
distributions before and after the outbreak of the pandemic and found that
traffic accidents have shifted in both location and time compared to previous
years. In addition, we have studied the road network characteristics in those
hotspot regions with the hope to understand the underlying cause of the hotspot
shifts.

【2】 U-Net-and-a-half: Convolutional network for biomedical image  segmentation using multiple expert-driven annotations

：Yichi Zhang,Jesper Kers,Clarissa A. Cassol,Joris J. Roelofs,Najia Idrees,Alik Farber,Samir Haroon,Kevin P. Daly,Suvranu Ganguli,Vipul C. Chitalia,Vijaya B. Kolachalama

【3】 Known Operator Learning and Hybrid Machine Learning in Medical Imaging  --- A Review of the Past, the Present, and the Future

：Andreas Maier,Harald Köstler,Marco Heisig,Patrick Krauss,Seung Hee Yang

【4】 A Survey of Machine Learning Techniques for Detecting and Diagnosing  COVID-19 from Imaging

：Aishwarza Panday,Muhammad Ashad Kabir,Nihad Karim Chowdhury

transcription-polymerase chain reaction (RT-PCR) test, many studies have
proposed machine learning techniques for detecting COVID-19 from medical
imaging. The purpose of this study is to systematically review, assess, and
synthesize research articles that have used different machine learning
techniques to detect and diagnose COVID-19 from chest X-ray and CT scan images.
A structured literature search was conducted in the relevant bibliographic
databases to ensure that the survey solely centered on reproducible and
high-quality research. We selected papers based on our inclusion criteria. In
this survey, we reviewed $98$ articles that fulfilled our inclusion criteria.
We have surveyed a complete pipeline of chest imaging analysis techniques
related to COVID-19, including data collection, pre-processing, feature
extraction, classification, and visualization. We have considered CT scans and
X-rays as both are widely used to describe the latest developments in medical
imaging to detect COVID-19. This survey provides researchers with valuable
insights into different machine learning techniques and their performance in
the detection and diagnosis of COVID-19 from chest imaging. At the end, the
challenges and limitations in detecting COVID-19 using machine learning
techniques and the future direction of research are discussed.

【5】 Automated Olfactory Bulb Segmentation on High Resolutional T2-Weighted  MRI