Learning Task Research Articles

Weakly Supervised AUC Optimization: A Unified Partial AUC Approach.

Since acquiring perfect supervision is usually difficult, real-world machine learning tasks often confront inaccurate, incomplete, or inexact supervision, collectively referred to as weak supervision. In this work, we present WSAUC, a unified framework for weakly supervised AUC optimization problems, which covers noisy label learning, positive-unlabeled learning, multi-instance learning, and semi-supervised learning scenarios. Within the WSAUC framework, we first frame the AUC optimization problems in various weakly supervised scenarios as a common formulation of minimizing the AUC risk on contaminated sets, and demonstrate that the empirical risk minimization problems are consistent with the true AUC. Then, we introduce a new type of partial AUC, specifically, the reversed partial AUC (rpAUC), which serves as a robust training objective for AUC maximization in the presence of contaminated labels. WSAUC offers a universal solution for AUC optimization in various weakly supervised scenarios by maximizing the empirical rpAUC. Theoretical and experimental results under multiple settings support the effectiveness of WSAUC on a range of weakly supervised AUC optimization tasks.

Vulnerability to semantic and phonological interference in normal aging and amnestic mild cognitive impairment (aMCI).

To determine whether the increased vulnerability to semantic interference previously observed in amnestic mild cognitive impairment (aMCI) is specifically associated with semantic material or if it also affects other types of material, suggesting generalized executive and inhibitory impairment. Seventy-two participants divided into two groups (33 aMCI, 39 normal control [NC]) matched for age and education were included. They completed a comprehensive neuropsychological examination, the French version of the Loewenstein Acevedo Scale for Semantic Interference and Learning (LASSI-L; semantic interference test), and a homologous experimental phonological test, the phonological interference and learning test. Independent sample t tests, mixed analysis of variance (ANOVA), and analysis of covariance (ANCOVA) on memory and interference scores were conducted to compare memory and interference in both conditions for both groups. For memory scores, results revealed significant main effects of group (NC > aMCI) and condition (semantic > phonological) and significant interactions (poorer performance in the semantic condition for aMCI). aMCI committed more phonological false recognition errors, were disproportionately more vulnerable to retroactive semantic interference, and showed a higher percentage of intrusion errors associated with proactive semantic interference than NC. To our knowledge, this is the first study to compare vulnerability to interference in aMCI and normal aging with two similarly designed semantic and phonological word list learning tasks. Taken together, our results suggest that aMCI present with broad difficulties in source memory and inhibition, but that impaired deep semantic processing results in additional semantic intrusion errors during proactive interference and impacts their ability to show good recall after an interference list (greater semantic retroactive interference). Results are discussed according to the level-of-processing and activation/monitoring theories. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Hilbert Curve Projection Distance for Distribution Comparison.

Distribution comparison plays a central role in many machine learning tasks like data classification and generative modeling. In this study, we propose a novel metric, called Hilbert curve projection (HCP) distance, to measure the distance between two probability distributions with low complexity. In particular, we first project two high-dimensional probability distributions using Hilbert curve to obtain a coupling between them, and then calculate the transport distance between these two distributions in the original space, according to the coupling. We show that HCP distance is a proper metric and is well-defined for probability measures with bounded supports. Furthermore, we demonstrate that the modified empirical HCP distance with the Lp cost in the d-dimensional space converges to its population counterpart at a rate of no more than O(n-1/2max{d,p}). To suppress the curse-of-dimensionality, we also develop two variants of the HCP distance using (learnable) subspace projections. Experiments on both synthetic and real-world data show that our HCP distance works as an effective surrogate of the Wasserstein distance with low complexity and overcomes the drawbacks of the sliced Wasserstein distance.

Construction of Practical Teaching Mode of Law Course Based on Multi-Mode and Low-Resource Language Learning

China's legal profession is a new profession emerging in the process of China's rule of law. Its development and refinement are increasing in accordance with the rapid social, economic and political development. However, as a lawyer, people need to learn real kung fu. For complex and changeable legal practice problems, people should use the simplest and most effective way to solve them. Therefore, it is necessary to study the curriculum of law specialty to achieve its purpose. It is also necessary to reform the traditional education methods, introduce modern scientific and technological means, and improve their effectiveness, so as to construct a practical teaching mode of Professional Courses in Law (PCL). For practical legal courses, it can be conducted through multi-mode low-resource language learning method. This paper aimed to study how to study the ability of students to obtain information, process information and analyze and process information based on intelligent multi-image feature fusion. Based on the investigation of teaching objectives and teachers' teaching requirements, and the analysis and arrangement of the investigation results, a multi-task learning system model based on image intelligent features was established. In this experiment, 494 valid questionnaires were used to analyze the current situation of PCL practical teaching. Among them, 10.32% and 65.59% of the students were very satisfied with the results of PCL practical teaching, and 18.02% and 59.31% of the students liked PCL practical teaching very much. In the survey on the influencing factors of PCL practical teaching activities, 61.34% of the students thought that the students did not know enough and were not enthusiastic enough. On the whole, the students are satisfied with the practical teaching arrangement of PCL, but from the actual situation, the effect of practical teaching is not very good, and the expected goal has not been achieved.

Acute Stress Effects on Statistical Learning and Episodic Memory.

Stress is widely considered to negatively impact hippocampal function, thus impairing episodic memory. However, the hippocampus is not merely the seat of episodic memory. Rather, it also (via distinct circuitry) supports statistical learning. On the basis of rodent work suggesting that stress may impair the hippocampal pathway involved in episodic memory while sparing or enhancing the pathway involved in statistical learning, we developed a behavioral experiment to investigate the effects of acute stress on both episodic memory and statistical learning in humans. Participants were randomly assigned to one of three conditions: stress (socially evaluated cold pressor) immediately before learning, stress ∼15 min before learning, or no stress. In the learning task, participants viewed a series of trial-unique scenes (allowing for episodic encoding of each image) in which certain scene categories reliably followed one another (allowing for statistical learning of associations between paired categories). Memory was assessed 24 hr later to isolate stress effects on encoding/learning rather than retrieval. We found modest support for our hypothesis that acute stress can amplify statistical learning: Only participants stressed ∼15 min in advance exhibited reliable evidence of learning across multiple measures. Furthermore, stress-induced cortisol levels predicted statistical learning retention 24 hr later. In contrast, episodic memory did not differ by stress condition, although we did find preliminary evidence that acute stress promoted memory for statistically predictable information and attenuated competition between statistical and episodic encoding. Together, these findings provide initial insights into how stress may differentially modulate learning processes within the hippocampus.

A Fully Test-time Training Framework for Semi-supervised Node Classification on Out-of-Distribution Graphs

Graph neural networks (GNNs) have shown great potential in representation learning for various graph tasks. However, the distribution shift between the training and test sets poses a challenge to the efficiency of GNNs. To address this challenge, HomoTTT proposes a fully test-time training framework for GNNs to enhance the model’s generalization capabilities for node classification tasks. Specifically, our proposed HomoTTT designs a homophily-based and parameter-free graph contrastive learning task with adaptive augmentation to guide the model’s adaptation during the test-time training, allowing the model to adapt for specific target data. In the inference stage, HomoTTT proposes to integrate the original GNN model and the adapted model after TTT using a homophily-based model selection method, which prevents potential performance degradation caused by unconstrained model adaptation. Extensive experimental results on six benchmark datasets demonstrate the effectiveness of our proposed framework. Additionally, the exploratory study further validates the rationality of the homophily-based graph contrastive learning task with adaptive augmentation and the homophily-based model selection designed in HomoTTT .

Federated Momentum Contrastive Clustering

Self-supervised representation learning and deep clustering are mutually beneficial to learn high-quality representations and cluster data simultaneously in centralized settings. However, it is not always feasible to gather large amounts of data at a central entity, considering data privacy requirements and computational resources. Federated Learning (FL) has been developed successfully to aggregate a global model while training on distributed local data, respecting the data privacy of edge devices. However, most FL research effort focuses on supervised learning algorithms. A fully unsupervised federated clustering scheme has not been considered in the existing literature. We present federated momentum contrastive clustering (FedMCC), a generic federated clustering framework that can not only cluster data automatically but also extract discriminative representations training from distributed local data over multiple users. In FedMCC, we demonstrate a two-stage federated learning paradigm where the first stage aims to learn differentiable instance embeddings and the second stage accounts for clustering data automatically. The experimental results show that FedMCC not only achieves superior clustering performance but also outperforms several existing federated self-supervised methods for linear evaluation and semi-supervised learning tasks. Additionally, FedMCC can easily be adapted to ordinary centralized clustering through what we call momentum contrastive clustering (MCC). We show that MCC achieves state-of-the-art clustering accuracy results in certain datasets such as STL-10 and ImageNet-10. We also present a method to reduce the memory footprint of our clustering schemes.

Deterioration in cognitive control related mPFC function underlying development of treatment resistance in early psychosis

One third of people with psychosis become antipsychotic treatment-resistant and the underlying mechanisms remain unclear. We investigated whether altered cognitive control function is a factor underlying development of treatment resistance. We studied 50 people with early psychosis at a baseline visit (mean < 2 years illness duration) and follow-up visit (1 year later), when 35 were categorized at treatment-responsive and 15 as treatment-resistant. Participants completed an emotion-yoked reward learning task that requires cognitive control whilst undergoing fMRI and MR spectroscopy to measure glutamate levels from Anterior Cingulate Cortex (ACC). Changes in cognitive control related activity (in prefrontal cortex and ACC) over time were compared between treatment-resistant and treatment-responsive groups and related to glutamate. Compared to treatment-responsive, treatment-resistant participants showed blunted activity in right amygdala (decision phase) and left pallidum (feedback phase) at baseline which increased over time and was accompanied by a decrease in medial Prefrontal Cortex (mPFC) activity (feedback phase) over time. Treatment-responsive participants showed a negative relationship between mPFC activity and glutamate levels at follow-up, no such relationship existed in treatment-resistant participants. Reduced activity in right amygdala and left pallidum at baseline was predictive of treatment resistance at follow-up (67% sensitivity, 94% specificity). The findings suggest that deterioration in mPFC function over time, a key cognitive control region needed to compensate for an initial dysfunction within a social-emotional network, is a factor underlying development of treatment resistance in early psychosis. An uncoupling between glutamate and cognitive control related mPFC function requires further investigation that may present a future target for interventions.

Time-dependent relations between emotion regulation, frustration, and metacognitive strategy use in technology-mediated learning

ABSTRACT Understanding how learners regulate their emotions and engage in metacognitive strategies is crucial for fostering self-regulated learning, particularly in technology-mediated learning. This study examines the temporal relationships between two emotion regulation (ER) strategies, reappraisal and suppression, frustration, and use of progress monitoring as metacognitive strategy, within the context of an educational game on financial literacy. The study involved 82 undergraduate students whose levels of frustration, progress monitoring behaviour, ER strategies were assessed at various points during the learning task. Findings revealed that the use of both reappraisal and suppression decreased during the learning task. Additionally, both ER strategies were negatively associated with frustration, although the relationship between reappraisal and frustration diminished over time. Frustration was negatively related to progress monitoring, indicating that effective emotion regulation can help maintain engagement in metacognitive strategies by keeping cognitive resources available. Notably, suppression and progress monitoring showed a positive relation that increased over time, highlighting the potential usefulness of suppression in extended learning tasks, despite its generally lower effectiveness compared to reappraisal. The results highlight the importance of considering temporal dynamics in the application of ER strategies during extended learning. Practical implications for the design of technology-mediated learning environments and educational interventions are proposed.

Guiding real-world reinforcement learning for in-contact manipulation tasks with Shared Control Templates

The requirement for a high number of training episodes has been a major limiting factor for the application of Reinforcement Learning (RL) in robotics. Learning skills directly on real robots requires time, causes wear and tear and can lead to damage to the robot and environment due to unsafe exploratory actions. The success of learning skills in simulation and transferring them to real robots has also been limited by the gap between reality and simulation. This is particularly problematic for tasks involving contact with the environment as contact dynamics are hard to model and simulate. In this paper we propose a framework which leverages a shared control framework for modeling known constraints defined by object interactions and task geometry to reduce the state and action spaces and hence the overall dimensionality of the reinforcement learning problem. The unknown task knowledge and actions are learned by a reinforcement learning agent by conducting exploration in the constrained environment. Using a pouring task and grid-clamp placement task (similar to peg-in-hole) as use cases and a 7-DoF arm, we show that our approach can be used to learn directly on the real robot. The pouring task is learned in only 65 episodes (16 min) and the grid-clamp placement task is learned in 75 episodes (17 min) with strong safety guarantees and simple reward functions, greatly alleviating the need for simulation.

Artificial Intelligence as a Dynamic Copilot in Entrepreneurship Education

How will instructors effectively manage the use of artificial intelligence (AI) with generative capabilities in entrepreneurship education? This paper introduces a framework to help instructors understand the different roles played by AI and learners as they move through different phases of a learning task. We describe the Artificial Intelligence in Entrepreneurship Education (AIEE) framework to outline the roles and relationships that instructors, students, and AI systems may play as part of a learning task in the classroom. We articulate key takeaways for instructors to recognize the responsibilities that are associated with different roles at different phases of a learning task.

Investigating the Impact of Machine Learning in Personalized Education Systems

Personalized education systems leverage robust capabilities of machine learning (ML). Tailoring learning experiences to individual needs preferences and abilities. This survey explores integration of ML in personalized education. It examines methodologies. It highlights applications impacts Future directions are discussed. Key ML techniques include supervised learning for performance prediction and feedback personalization. Unsupervised learning for identifying learning styles. Reinforcement learning.Personalized education systems leverage robust capabilities of machine learning (ML). Tailoring learning experiences to individual needs preferences and abilities. This survey explores integration of ML in personalized education. It examines methodologies. Highlighting applications. Impacts and future directions are discussed. Key ML techniques include supervised learning for performance prediction, feedback personalization unsupervised learning for identifying learning styles, reinforcement learning. Optimizing content sequences. Deep learning for tasks like automated grading. Intelligent tutoring included. Applications such as intelligent tutoring systems. Adaptive learning platforms. Learning analytics. Personalized learning paths and automated assessment systems discussed. The impact of ML-driven personalized education systems is profound. They enhance engagement. Motivation. Academic performance and reduce dropout rates. Challenges such as data privacy algorithmic bias, scalability and resistance to change are also addressed. Future research directions include interdisciplinary collaboration. Advancements in natural language processing are essential. Enhanced personalization algorithms are needed. Long-term studies. Considered crucial. Areas for ongoing work. Key Words: Machine learning, Personalized education, Adaptive learning, Intelligent tutoring systems, Predictive analytics, Recommendation systems, Ethical considerations, Student engagement, Educational technology

Advanced facial recognition with LBP-URIGL hybrid descriptors

AbstractFacial recognition technology is transformative in security and human-machine interaction, reshaping societal interactions. Robust descriptors, essential for high precision in machine learning tasks like recognition and recall, are integral to this transformation. This paper presents a hybrid model enhancing local binary pattern descriptors for facial representation. By integrating rotation-invariant local binary pattern with uniform rotation-invariant grey-level co-occurrence, employing linear discriminant analysis for feature space optimization, and utilizing an artificial neural network for classification, the model achieves exceptional accuracy rates of 100% for Olivetti Research Laboratory, 99.98% for Maastricht University Computer Vision Test, and 99.17% for Extended Yale B, surpassing traditional methods significantly.

Microstructural differences in the cingulum and the inferior longitudinal fasciculus are associated with (extinction) learning

Cognitive functions, such as learning and memory processes, depend on effective communication between brain regions which is facilitated by white matter tracts (WMT). We investigated the microstructural properties and the contribution of WMT to extinction learning and memory in a predictive learning task. Forty-two healthy participants completed an extinction learning paradigm without a fear component. We examined differences in microstructural properties using diffusion tensor imaging to identify underlying neural connectivity and structural correlates of extinction learning and their potential implications for the renewal effect. Participants with good acquisition performance exhibited higher fractional anisotropy (FA) in WMT including the bilateral inferior longitudinal fasciculus (ILF) and the right temporal part of the cingulum (CNG). This indicates enhanced connectivity and communication between brain regions relevant to learning and memory resulting in better learning performance. Our results suggest that successful acquisition and extinction performance were linked to enhanced structural connectivity. Lower radial diffusivity (RD) in the right ILF and right temporal part of the CNG was observed for participants with good acquisition learning performance. This observation suggests that learning difficulties associated with increased RD may potentially be due to less myelinated axons in relevant WMT. Also, participants with good acquisition performance were more likely to show a renewal effect. The results point towards a potential role of structural integrity in extinction-relevant WMT for acquisition and extinction.

Discrete Structural Design Synthesis: A Hierarchical-Inspired Deep Reinforcement Learning Approach Considering Topological and Parametric Actions

Abstract Structural design synthesis considering discrete elements can be formulated as a sequential decision process solved using deep reinforcement learning, as shown in prior work. By modeling structural design synthesis as a Markov decision process (MDP), the states correspond to specific structural designs, the discrete actions correspond to specific design alterations, and the rewards are related to the improvement in the altered design’s performance with respect to the design objective and specified constraints. Here, the MDP action definition is extended by integrating parametric design grammars that further enable the design agent to not only alter a given structural design’s topology, but also its element parameters. In considering topological and parametric actions, both the dimensionality of the state and action space and the diversity of the action types available to the agent in each state significantly increase, making the overall MDP learning task more challenging. Hence, this paper also addresses discrete design synthesis problems with large state and action spaces by significantly extending the network architecture. Specifically, a hierarchical-inspired deep neural network architecture is developed to allow the agent to learn the type of action, topological or parametric, to apply, thus reducing the complexity of possible action choices in a given state. This extended framework is applied to the design synthesis of planar structures considering both discrete elements and cross-sectional areas, and it is observed to adeptly learn policies that synthesize high performing design solutions.

The effect of feedback timing on category learning and feedback processing in younger and older adults.

Corrective feedback can be received immediately after an action or with a temporal delay. Neuroimaging studies suggest that immediate and delayed feedback are processed by the striatum and medial temporal lobes (MTL), respectively. Age-related changes in the striatum and MTL may influence the efficiency of feedback-based learning in older adults. The current study leverages event-related potentials (ERPs) to evaluate age-related differences in immediate and delayed feedback processing and consequences for learning. The feedback-related negativity (FRN) captures activity in the frontostriatal circuit while the N170 is hypothesized to reflect MTL activation. 18 younger (Myears = 24.4) and 20 older (Myears = 65.5) adults completed learning tasks with immediate and delayed feedback. For each group, learning outcomes and ERP magnitudes were evaluated across timing conditions. Younger adults learned better than older adults in the immediate timing condition. This performance difference was associated with a typical FRN signature in younger but not older adults. For older adults, impaired processing of immediate feedback in the striatum may have negatively impacted learning. Conversely, learning was comparable across groups when feedback was delayed. For both groups, delayed feedback was associated with a larger magnitude N170 relative to immediate feedback, suggesting greater MTL activation. Delaying feedback may increase MTL involvement and, for older adults, improve category learning. Age-related neural changes may differentially affect MTL- and striatal-dependent learning. Future research can evaluate the locus of age-related learning differences and how feedback can be manipulated to optimize learning across the lifespan.

Phonological learning is asymmetrical between prefixes and suffixes

Phonological theories typically treat prefixes and suffixes as if phonological processes apply uniformly to both. However, previous studies have found clear asymmetries between the behavior of both affixes. We argue that this is due to phonological processes being easier to learn in suffix position rather than prefix position. To test this, an artificial language learning task was used to evaluate whether phonological learning is symmetrical between affixes. Results pattern in the direction of the hypothesis, namely, that phonological learning is asymmetrically facilitated in suffix position over prefix position, providing support to general cognitive mechanisms being responsible for strong cross-linguistic tendencies.

Does learning from language family help? A case study on a low-resource question-answering task

Abstract Multilingual pre-trained models make it possible to develop natural language processing (NLP) applications for low-resource languages (LRLs) using the model of resource-rich languages (RRLs). However, the structural characteristics of the target languages can impact task-specific learning. In this paper, we investigate the influence of structural diversities of languages on the system’s overall performance. Specifically, we propose a customized approach to leverage task-specific data of low-resource language families via transfer learning from RRL. Our findings are based on question-answering tasks using the XLM-R, mBERT, and IndicBERT transformer models and Indic languages (Hindi, Bengali, and Telugu). On the XQuAD-Hindi dataset, the few-shot learning using Bengali improves the benchmark mBERT (F1/EM) score by +(10.86/7.87) and XLM-R score by +(3.84/4.42). Few-shot learning using Telugu has also improved the mBERT score by +(10.42/7.36) and +(3.04/2.72) for XLM-R. In addition, our model has demonstrated benchmark-compatible performance in a zero-shot setup with single-epoch task learning. This approach can be adapted for other NLP tasks for LRLs.

Improving job matching with deep learning-based hyper-personalization

This study introduces a novel approach to streamline the recruitment process, benefiting both employers and job seekers. It leverages real-time personality-based classification to match candidates with the most suitable roles in a scalable and precise manner. This is achieved through machine learning-driven hyper-personalization, employing deep learning models to create a predictive language model. The study encompasses two key tasks: binary classification, distinguishing sentences containing soft skills (1) from those that do not (0), and multi-class classification, categorizing positive sentences into five classes based on Big Five personality traits. The research involved a series of experiments. Initially, multiple machine learning algorithms were employed to establish baseline models. Subsequently, the study investigated the impact of deep learning versus these baseline models. The results demonstrated an accuracy of 0.79% and 0.68% for binary classification tasks, and 0.79% and 0.60% for multi-class classification tasks, using Support Vector Machines in the machine learning task, and Bidirectional Long Short-Term Memory in the deep learning task, respectively. This approach showcases promise in revolutionizing the job matching process, offering a more efficient and accurate means of connecting individuals with their ideal employment opportunities based on their unique soft skills and personality traits.

Contrastive and uncertainty-aware nuclei segmentation and classification

Nuclei segmentation and classification play a crucial role in pathology diagnosis, enabling pathologists to analyze cellular characteristics accurately. Overlapping cluster nuclei, misdetection of small-scale nuclei, and pleomorphic nuclei-induced misclassification have always been major challenges in the nuclei segmentation and classification tasks. To this end, we introduce an auxiliary task of nuclei boundary-guided contrastive learning to enhance the representativeness and discriminative power of visual features, particularly for addressing the challenge posed by the unclear contours of adherent nuclei and small nuclei. In addition, misclassifications resulting from pleomorphic nuclei often exhibit low classification confidence, indicating a high level of uncertainty. To mitigate misclassification, we capitalize on the characteristic clustering of similar cells to propose a locality-aware class embedding module, offering a regional perspective to capture category information. Moreover, we address uncertain classification in densely aggregated nuclei by designing a top-k uncertainty attention module that leverages deep features to enhance shallow features, thereby improving the learning of contextual semantic information. We demonstrate that the proposed network outperforms the off-the-shelf methods in both nuclei segmentation and classification experiments, achieving the state-of-the-art performance.