Multi-modality Fluorodeoxyglucose (FDG) positron emission tomography / computed tomography (PET/CT) has been routinely used in the assessment of common cancers, such as lung cancer, lymphoma, and melanoma. This is mainly attributed to the fact that PET/CT combines the high sensitivity for tumor detection of PET and anatomical information from CT. In PET/CT image assessment, automatic tumor segmentation is an important step, and in recent years, deep learning based methods have become the state-of-the-art. Unfortunately, existing methods tend to over-segment the tumor regions and include regions such as the normal high uptake organs, inflammation, and other infections. In this study, we introduce a false positive reduction network to overcome this limitation. We firstly introduced a self-supervised pre-trained global segmentation module to coarsely delineate the candidate tumor regions using a self-supervised pre-trained encoder. The candidate tumor regions were then refined by removing false positives via a local refinement module. Our experiments with the MICCAI 2022 Automated Lesion Segmentation in Whole-Body FDG-PET/CT (AutoPET) challenge dataset showed that our method achieved a dice score of 0.9324 with the preliminary testing data and was ranked 1st place in dice on the leaderboard. Our method was also ranked in the top 7 methods on the final testing data, the final ranking will be announced during the 2022 MICCAI AutoPET workshop.

Since the development of writing 5000 years ago, human-generated data gets produced at an ever-increasing pace. Classical archival methods aimed at easing information retrieval. Nowadays, archiving is not enough anymore. The amount of data that gets generated daily is beyond human comprehension, and appeals for new information retrieval strategies. Instead of referencing every single data piece as in traditional archival techniques, a more relevant approach consists in understanding the overall ideas conveyed in data flows. To spot such general tendencies, a precise comprehension of the underlying data generation mechanisms is required. In the rich literature tackling this problem, the question of information interaction remains nearly unexplored. First, we investigate the frequency of such interactions. Building on recent advances made in Stochastic Block Modelling, we explore the role of interactions in several social networks. We find that interactions are rare in these datasets. Then, we wonder how interactions evolve over time. Earlier data pieces should not have an everlasting influence on ulterior data generation mechanisms. We model this using dynamic network inference advances. We conclude that interactions are brief. Finally, we design a framework that jointly models rare and brief interactions based on Dirichlet-Hawkes Processes. We argue that this new class of models fits brief and sparse interaction modelling. We conduct a large-scale application on Reddit and find that interactions play a minor role in this dataset. From a broader perspective, our work results in a collection of highly flexible models and in a rethinking of core concepts of machine learning. Consequently, we open a range of novel perspectives both in terms of real-world applications and in terms of technical contributions to machine learning.

Distributed deep learning frameworks such as split learning provide great benefits with regards to the computational cost of training deep neural networks and the privacy-aware utilization of the collective data of a group of data-holders. Split learning, in particular, achieves this goal by dividing a neural network between a client and a server so that the client computes the initial set of layers, and the server computes the rest. However, this method introduces a unique attack vector for a malicious server attempting to steal the client's private data: the server can direct the client model towards learning any task of its choice, e.g. towards outputting easily invertible values. With a concrete example already proposed (Pasquini et al., CCS '21), such training-hijacking attacks present a significant risk for the data privacy of split learning clients. In this paper, we propose SplitGuard, a method by which a split learning client can detect whether it is being targeted by a training-hijacking attack or not. We experimentally evaluate our method's effectiveness, compare it with potential alternatives, and discuss in detail various points related to its use. We conclude that SplitGuard can effectively detect training-hijacking attacks while minimizing the amount of information recovered by the adversaries.

LiDAR sensor is essential to the perception system in autonomous vehicles and intelligent robots. To fulfill the real-time requirements in real-world applications, it is necessary to efficiently segment the LiDAR scans. Most of previous approaches directly project 3D point cloud onto the 2D spherical range image so that they can make use of the efficient 2D convolutional operations for image segmentation. Although having achieved the encouraging results, the neighborhood information is not well-preserved in the spherical projection. Moreover, the temporal information is not taken into consideration in the single scan segmentation task. To tackle these problems, we propose a novel approach to semantic segmentation for LiDAR sequences named Meta-RangeSeg, where a new range residual image representation is introduced to capture the spatial-temporal information. Specifically, Meta-Kernel is employed to extract the meta features, which reduces the inconsistency between the 2D range image coordinates input and 3D Cartesian coordinates output. An efficient U-Net backbone is used to obtain the multi-scale features. Furthermore, Feature Aggregation Module (FAM) strengthens the role of range channel and aggregates features at different levels. We have conducted extensive experiments for performance evaluation on SemanticKITTI and SemanticPOSS. The promising results show that our proposed Meta-RangeSeg method is more efficient and effective than the existing approaches. Our full implementation is publicly available at https://github.com/songw-zju/Meta-RangeSeg .

Text classifiers are applied at scale in the form of one-size-fits-all solutions. Nevertheless, many studies show that classifiers are biased regarding different languages and dialects. When measuring and discovering these biases, some gaps present themselves and should be addressed. First, ``Does language, dialect, and topical content vary across geographical regions?'' and secondly ``If there are differences across the regions, do they impact model performance?''. We introduce a novel dataset called GeoOLID with more than 14 thousand examples across 15 geographically and demographically diverse cities to address these questions. We perform a comprehensive analysis of geographical-related content and their impact on performance disparities of offensive language detection models. Overall, we find that current models do not generalize across locations. Likewise, we show that while offensive language models produce false positives on African American English, model performance is not correlated with each city's minority population proportions. Warning: This paper contains offensive language.

Document-level information extraction (IE) tasks have recently begun to be revisited in earnest using the end-to-end neural network techniques that have been successful on their sentence-level IE counterparts. Evaluation of the approaches, however, has been limited in a number of dimensions. In particular, the precision/recall/F1 scores typically reported provide few insights on the range of errors the models make. We build on the work of Kummerfeld and Klein (2013) to propose a transformation-based framework for automating error analysis in document-level event and (N-ary) relation extraction. We employ our framework to compare two state-of-the-art document-level template-filling approaches on datasets from three domains; and then, to gauge progress in IE since its inception 30 years ago, vs. four systems from the MUC-4 (1992) evaluation.

Distributional shift, or the mismatch between training and deployment data, is a significant obstacle to the usage of machine learning in high-stakes industrial applications, such as autonomous driving and medicine. This creates a need to be able to assess how robustly ML models generalize as well as the quality of their uncertainty estimates. Standard ML baseline datasets do not allow these properties to be assessed, as the training, validation and test data are often identically distributed. Recently, a range of dedicated benchmarks have appeared, featuring both distributionally matched and shifted data. Among these benchmarks, the Shifts dataset stands out in terms of the diversity of tasks as well as the data modalities it features. While most of the benchmarks are heavily dominated by 2D image classification tasks, Shifts contains tabular weather forecasting, machine translation, and vehicle motion prediction tasks. This enables the robustness properties of models to be assessed on a diverse set of industrial-scale tasks and either universal or directly applicable task-specific conclusions to be reached. In this paper, we extend the Shifts Dataset with two datasets sourced from industrial, high-risk applications of high societal importance. Specifically, we consider the tasks of segmentation of white matter Multiple Sclerosis lesions in 3D magnetic resonance brain images and the estimation of power consumption in marine cargo vessels. Both tasks feature ubiquitous distributional shifts and a strict safety requirement due to the high cost of errors. These new datasets will allow researchers to further explore robust generalization and uncertainty estimation in new situations. In this work, we provide a description of the dataset and baseline results for both tasks.

Given a discriminating neural network, the problem of fairness improvement is to systematically reduce discrimination without significantly scarifies its performance (i.e., accuracy). Multiple categories of fairness improving methods have been proposed for neural networks, including pre-processing, in-processing and post-processing. Our empirical study however shows that these methods are not always effective (e.g., they may improve fairness by paying the price of huge accuracy drop) or even not helpful (e.g., they may even worsen both fairness and accuracy). In this work, we propose an approach which adaptively chooses the fairness improving method based on causality analysis. That is, we choose the method based on how the neurons and attributes responsible for unfairness are distributed among the input attributes and the hidden neurons. Our experimental evaluation shows that our approach is effective (i.e., always identify the best fairness improving method) and efficient (i.e., with an average time overhead of 5 minutes).

The Medical Internet of Things, a recent technological advancement in medicine, is incredibly helpful in providing real-time monitoring of health metrics. This paper presents a non-invasive IoT system that tracks patients' emotions, especially those with an autism spectrum disorder. With a few affordable sensors and cloud computing services, the individual's heart rates are monitored and analyzed to study the effects of change in sweat and heartbeats per minute for different emotions. Under normal resting conditions of the individual, the proposed system could detect the right emotion using machine learning algorithms with a performance of up to 92% accuracy. The result of the proposed approach is comparable with the state-of-the-art solutions in medical IoT.

Research articles are being shared in increasing numbers on multiple online platforms. Although the scholarly impact of these articles has been widely studied, the online interest determined by how long the research articles are shared online remains unclear. Being cognizant of how long a research article is mentioned online could be valuable information to the researchers. In this paper, we analyzed multiple social media platforms on which users share and/or discuss scholarly articles. We built three clusters for papers, based on the number of yearly online mentions having publication dates ranging from the year 1920 to 2016. Using the online social media metrics for each of these three clusters, we built machine learning models to predict the long-term online interest in research articles. We addressed the prediction task with two different approaches: regression and classification. For the regression approach, the Multi-Layer Perceptron model performed best, and for the classification approach, the tree-based models performed better than other models. We found that old articles are most evident in the contexts of economics and industry (i.e., patents). In contrast, recently published articles are most evident in research platforms (i.e., Mendeley) followed by social media platforms (i.e., Twitter).