Despite significant advancements in deep classification techniques and in-lab automatic optical inspection (AOI) models for long-tailed or highly imbalanced data, applying these approaches to real-world IC defect classification tasks remains challenging. This difficulty stems from two primary factors. First, real-world conditions, such as the high yield-rate requirements in the IC industry, result in data distributions that are far more skewed than those found in general public imbalanced datasets. Consequently, classifiers designed for open imbalanced datasets often fail to perform effectively in real-world scenarios. Second, real-world samples exhibit a mix of class-specific attributes (e.g., defect types) and class-agnostic, domain-related features (e.g., design characteristics of product lines). This complexity adds significant difficulty to the classification process, particularly for highly imbalanced datasets. To address these challenges, this paper introduces the IC-Defect-14 dataset, a large, highly imbalanced IC defect image dataset sourced from AOI systems deployed in real-world IC production lines. This dataset is characterized by its unique “intra-class clusters” property, which presents two major challenges: large intra-class diversity and high inter-class similarity. These characteristics, rarely found simultaneously in existing public datasets, significantly degrade the performance of current state-of-the-art classifiers for highly imbalanced data. To tackle this challenge, we propose the Regional Channel Attention-based Multi-Expert Network (ReCAME-Net). This network follows a multi-expert classifier framework and integrates a regional channel attention module, metric learning losses, a hard category mining strategy, and a knowledge distillation procedure. Extensive experimental evaluations demonstrate that ReCAME-Net outperforms previous state-of-the-art models on the IC-Defect-14 dataset while maintaining comparable performance and competitiveness on general public datasets. Our resources can be found at https://github.com/YoursEver/ReCAME-Net.

This study aimed to examine the effects of collaborative anchoring on the development of digital curation skills among nursing college learners engaged in healthcare animation design activities. A cohort of thirty participants was recruited and evenly distributed into six groups to participate in three distinct animation design sessions, each incorporating different instructional interventions. Initially, participants engaged in a face-to-face discussion, during which they reviewed and critiqued the strengths and weaknesses of an assigned healthcare animation video. Subsequently, they participated in an online collaborative anchoring session utilizing a cloud-based application, SyncSketch, to critique another animation video within a structured digital environment. The final session involved anonymous online collaborative anchoring, wherein the same task was conducted, but all feedback was provided anonymously. A six-genre collaborative anchor framework was devised to analyze discrepancies in anchoring across the three instructional sessions. Additionally, a debugging task was administered to assess variations in digital curation skills following the experiment. The findings indicated that the use of SyncSketch's timeline anchors significantly enhanced participants' focus and engagement, resulting in more precise and targeted feedback compared to traditional face-to-face discussions. Furthermore, the anonymous online collaborative anchoring fostered a more equitable peer review dynamic, ensuring that all participants could contribute meaningfully. The study concludes by discussing the implications of integrating collaborative anchoring into nursing education as a means to enhance digital curation skills.

The use of mobile integrated access and backhaul (mIAB) is promising due to its high cost-effectiveness and fast deployment. However, the study of the deployment optimization for mIAB nodes is still incomplete, especially for air-ground integrated networks. To fill the gap, this paper aims to develop non-model-based deployment approaches for both unconstrained and constrained deployment problems. To this end, approaches with the zeroth order optimization techniques are proposed for solving deployment problems along with the use of deep neural network (DNN) surrogate models to help predicting network performance. In addition, effective training and design approaches for DNN surrogate models are provided by using data augmentation and active learning techniques. Simulation results show that our proposed approaches are effective and can outperform reference schemes.

Anomaly detection (AD) plays an important role in numerous applications. In this paper, we focus on two understudied aspects of AD that are critical for integration into real-world applications. First, most AD methods cannot incorporate labeled data that are often available in practice in small quantities and can be crucial to achieve high accuracy. Second, most AD methods are not interpretable, a bottleneck that prevents stakeholders from understanding the reason behind the anomalies. In this paper, we propose a novel AD framework, DIAD, that adapts a white-box model class, Generalized Additive Models, to detect anomalies using a partial identification objective which naturally handles noisy or heterogeneous features. DIAD can incorporate a small amount of labeled data to further boost AD performances in semi-supervised settings. We demonstrate the superiority of DIAD compared to previous work in both unsupervised and semi-supervised settings on multiple datasets. We also present explainability capabilities of DIAD, on its rationale behind predicting certain samples as anomalies.

Despite the success of machine learning applications in science, industry, and society in general, many approaches are known to be non-robust, often relying on spurious correlations to make predictions. Spuriousness occurs when some features correlate with labels but are not causal; relying on such features prevents models from generalizing to unseen environments where such correlations break. In this work, we focus on image classification and propose two data generation processes to reduce spuriousness. Given human annotations of the subset of the features responsible (causal) for the labels (e.g. bounding boxes), we modify this causal set to generate a surrogate image that no longer has the same label (i.e. a counterfactual image). We also alter non-causal features to generate images still recognized as the original labels, which helps to learn a model invariant to these features. In several challenging datasets, our data generations outperform state-of-the-art methods in accuracy when spurious correlations break, and increase the saliency focus on causal features providing better explanations.

Generalized additive models (GAMs) have become a leading model class for interpretable machine learning. However, there are many algorithms for training GAMs, and these can learn different or even contradictory models, while being equally accurate. Which GAM should we trust? In this paper, we quantitatively and qualitatively investigate a variety of GAM algorithms on real and simulated datasets. We find that GAMs with high feature sparsity (only using a few variables to make predictions) can miss patterns in the data and be unfair to rare subpopulations. Our results suggest that inductive bias plays a crucial role in what interpretable models learn and that tree-based GAMs represent the best balance of sparsity, fidelity and accuracy and thus appear to be the most trustworthy GAM models.

Models which estimate main effects of individual variables alongside interaction effects have an identifiability challenge: effects can be freely moved between main effects and interaction effects without changing the model prediction. This is a critical problem for interpretability because it permits “contradictory" models to represent the same function. To solve this problem, we propose pure interaction effects: variance in the outcome which cannot be represented by any subset of features. This definition has an equivalence with the Functional ANOVA decomposition. To compute this decomposition, we present a fast, exact algorithm that transforms any piecewise-constant function (such as a tree-based model) into a purified, canonical representation. We apply this algorithm to Generalized Additive Models with interactions trained on several datasets and show large disparity, including contradictions, between the apparent and the purified effects. These results underscore the need to specify data distributions and ensure identifiability before interpreting model parameters.

Imagine a patient in critical condition. What and when should be measured to forecast detrimental events, especially under the budget constraints? We answer this question by deep reinforcement learning (RL) that jointly minimizes the measurement cost and maximizes predictive gain, by scheduling strategically-timed measurements. We learn our policy to be dynamically dependent on the patient’s health history. To scale our framework to exponentially large action space, we distribute our reward in a sequential setting that makes the learning easier. In our simulation, our policy outperforms heuristic-based scheduling with higher predictive gain and lower cost. In a real-world ICU mortality prediction task (MIMIC3), our policies reduce the total number of measurements by 31% or improve predictive gain by a factor of 3 as compared to physicians, under the off-policy policy evaluation.

Both cameras and IoT devices have their particular capabilities in tracking moving objects. Their correlations are, however, unclear. In this work, we consider using a drone to track ground objects. We demonstrate how to retrieve IoT data from devices, which are attached on human objects, and correctly tag them on the human objects captured by a drone view. This is the first work correlating IoT data and computer vision from a drone camera. Potential applications of this work include aerial surveillance, people tracking, and intelligent human-drone interaction.

In this paper, we designed, implemented, and evaluated the performance of a Packet-in packet buffer system in an SDN bare metal commodity switch and Open vSwitch. In our approach, only the first packet of a new flow needs to be sent to the SDN controller via a Packet-in packet and all subsequent packets can be temporarily stored in the Packet-in packet buffer system until the arrival of the flow rule. Our experimental results show that our approach effectively prevents packets of a new flow from being dropped when they pass an SDN switch, significantly reduces the switch CPU usage, greatly reduces the control plane bandwidth usage, and greatly reduces the Packet-in packet processing load imposed on the SDN controller.

Mental disorders are currently affecting millions of people from different cultures, age groups and geographic regions. The challenge of mental disorders is that they are difficult to detect on suffering patients, thus presenting an alarming number of undetected cases and misdiagnosis. In this paper, we aim at building predictive models that leverage language and behavioral patterns, used particularly in social media, to determine whether a user is suffering from two cases of mental disorder. These predictive models are made possible by employing a novel data collection process, coined as Subconscious Crowdsourcing, which helps to collect a faster and more reliable dataset of patients. Our experiments suggest that extracting specific language patterns and social interaction features from reliable patient datasets can greatly contribute to further analysis and detection of mental disorders.

Mental illnesses rank as some of the most disabling conditions, affecting millions of people, across the globe. In general, the main challenge of mental disorders is that they remain difficult to detect on suffering patients. In an online environment, the challenge extends to the collection of patients data and the implementation of proper algorithms to assist in the detection of such illnesses. In this paper, we propose a novel data collection mechanism and build predictive models that leverage language and behavioral patterns, used particularly on Twitter, to determine whether a user is suffering from a mental disorder. After training the predictive models, they are further pre-trained to serve as the backend for our demonstration, MIDAS. MIDAS offers an analytics web-service to explore several characteristics pertaining to user's linguistic and behavioral patterns on social media, with respect to mental illnesses.

Opinionated user-generated content has been increasingly flooding the internet since the rise of the Web 2.0. Many of this content is generated by the occurrence of different events varying in time, scale and location. In recent years there has been a growing interest in having a deeper understanding of these events and how the public reacts to them. Towards this goal there is a constant development in areas such as opinion mining. Nevertheless these methods alone are insufficient to provide a greater insight regarding several event related behaviors. In this demonstration we present a time based visualization platform for analyzing events, focusing specifically on the emotional transition generated by their occurrence, to value the impact they have over society.

Parallelism is one promising solution to accelerating the test pattern generation (TPG) process, several recent works also show that parallel TPG can reduce the test pattern count. However, today's parallel TPG's are mostly non-deterministic, i.e., the generated test set is timing and resource dependent, this complicates the debug process and may degrade the user experience. In this paper, we propose a multi-threading parallel test pattern generator that is both deterministic and incurs zero test inflation. Called SDC-TPG, the proposed parallel TPG relies on synchronized dynamic compaction (SDC) to generate the same test pattern set as the conventional serial TPG with dynamic compaction regardless of the thread timing and the thread count. Furthermore, an early primary fault TPG strategy is proposed to reduce the thread idle times and improve the speedup. Simulation results show that SDC-TPG achieves an average speedup of six with eight threads.

This paper proposes a condition-based algorithm for H.264/AVC 4times4 intra prediction. Exploiting high correlation existed in neighboring intra prediction modes, we propose the three conditions to skip the less possible candidates in doing intra4times4 block mode decision. When compared to the 9 prediction modes in the full search algorithm, the proposed algorithm can complete a 4times4 intra prediction using 4.4 prediction modes operation in average. The simulation result shows that the proposed algorithm can reduce computational complexity up to 44% at the cost of less than 0.1 dB PSNR loss in average

In this paper, semi-supervised classification concept is applied to Discriminant Analysis Feature Extraction (DAFE) and Nonparametric Weighted Feature Extraction (NWFE). The proposed semi-supervised DAFE and NWFE use the information of both labeled and unlabeled data in an iterative process. Experimental results of hyperspectral data show that the proposed feature extraction methods can improve the classification performance significantly with limited training samples.