Machine learning with applications最新文献

Disinformation - the deliberate spread of false or misleading information poses a significant threat to our society by undermining trust, exacerbating polarization, and manipulating public opinion. With the rapid advancement of artificial intelligence and the growing prominence of large language models (LLMs) such as ChatGPT, new avenues for the dissemination of disinformation are emerging. This review paper explores the potential of LLMs to initiate the generation of multi-media disinformation, encompassing text, images, audio, and video. We begin by examining the capabilities of LLMs, highlighting their potential to create compelling, context-aware content that can be weaponized for malicious purposes. Subsequently, we examine the nature of disinformation and the various mechanisms through which it spreads in the digital landscape. Utilizing these advanced models, malicious actors can automate and scale up disinformation effectively. We describe a theoretical pipeline for creating and disseminating disinformation on social media. Existing interventions to combat disinformation are also reviewed. While these efforts have shown success, we argue that they need to be strengthened to effectively counter the escalating threat posed by LLMs. Digital platforms have, unfortunately, enabled malicious actors to extend the reach of disinformation. The advent of LLMs poses an additional concern as they can be harnessed to significantly amplify the velocity, variety, and volume of disinformation. Thus, this review proposes augmenting current interventions with AI tools like LLMs, capable of assessing information more swiftly and comprehensively than human fact-checkers. This paper illuminates the dark side of LLMs and highlights their potential to be exploited as disinformation dissemination tools.

This paper presents an automated machine learning framework designed to assist hydrologists in detecting anomalies in time series data generated by sensors in a research watershed in the northeastern United States critical zone. The framework specifically focuses on identifying peak-pattern anomalies, which may arise from sensor malfunctions or natural phenomena. However, the use of classification methods for anomaly detection poses challenges, such as the requirement for labeled data as ground truth and the selection of the most suitable deep learning model for the given task and dataset. To address these challenges, our framework generates labeled datasets by injecting synthetic peak patterns into synthetically generated time series data and incorporates an automated hyperparameter optimization mechanism. This mechanism generates an optimized model instance with the best architectural and training parameters from a pool of five selected models, namely Temporal Convolutional Network (TCN), InceptionTime, MiniRocket, Residual Networks (ResNet), and Long Short-Term Memory (LSTM). The selection is based on the user’s preferences regarding anomaly detection accuracy and computational cost. The framework employs Time-series Generative Adversarial Networks (TimeGAN) as the synthetic dataset generator. The generated model instances are evaluated using a combination of accuracy and computational cost metrics, including training time and memory, during the anomaly detection process. Performance evaluation of the framework was conducted using a dataset from a watershed, demonstrating consistent selection of the most fitting model instance that satisfies the user’s preferences.

Convolutional Neural Networks (CNNs) have shown remarkable prowess in detecting P300, an Event-Related Potential (ERP) crucial in Brain–Computer Interfaces (BCIs). Researchers persistently seek simple and efficient CNNs for P300 detection, exemplified by models like DeepConvNet, EEGNet, and SepConv1D. Noteworthy progress has been made, manifesting in reducing parameters from millions to hundreds while sustaining state-of-the-art performance. However, achieving further simplification or performance improvement beyond SepConv1D appears challenging due to inherent oversimplification. This study explores landmark CNNs and P300 data with the aid of Explainable AI, proposing a simpler yet superior-performing CNN architecture which incorporates (1) precise separable convolution for feature extraction of P300 data, (2) adaptive activation function tailored for P300 data, and (3) customized large learning rate schedules for training P300 data. Termed the Minimalist CNN for P300 detection (P300MCNN), this novel model is characterized by its requirement of the fewest filters and epochs to date, concurrently achieving best performance in cross-subject P300 detection. P300MCNN not only introduces groundbreaking concepts for CNN architectures in P300 detection but also showcases the importance of Explainable AI in demystifying the “black box” design of CNNs.

ChatGPT, a chatbot developed by OpenAI, has gained widespread popularity and media attention since its release in November 2022. However, little hard evidence is available regarding its perception in various sources. In this paper, we analyze over 300,000 tweets and more than 150 scientific papers to investigate how ChatGPT is perceived and discussed. Our findings show that ChatGPT is generally viewed as of high quality, with positive sentiment and emotions of joy dominating social media. Its perception has slightly decreased since its debut, however, with joy decreasing and (negative) surprise on the rise, and it is perceived more negatively in languages other than English. In recent scientific papers, ChatGPT is characterized as a great opportunity across various fields including the medical domain, but also as a threat concerning ethics and receives mixed assessments for education. Our comprehensive meta-analysis of ChatGPT’s perception after 2.5 months since its release can contribute to shaping the public debate and informing its future development. We make our data available.¹

Sports betting’s recent federal legalisation in the USA coincides with the golden age of machine learning. If bettors can leverage data to reliably predict the probability of an outcome, they can recognise when the bookmaker’s odds are in their favour. As sports betting is a multi-billion dollar industry in the USA alone, identifying such opportunities could be extremely lucrative. Many researchers have applied machine learning to the sports outcome prediction problem, generally using accuracy to evaluate the performance of predictive models. We hypothesise that for the sports betting problem, model calibration is more important than accuracy. To test this hypothesis, we train models on NBA data over several seasons and run betting experiments on a single season, using published odds. We show that using calibration, rather than accuracy, as the basis for model selection leads to greater returns, on average (return on investment of +34.69% versus -35.17%) and in the best case (+36.93% versus +5.56%). These findings suggest that for sports betting (or any probabilistic decision-making problem), calibration is a more important metric than accuracy. Sports bettors who wish to increase profits should therefore select their predictive model based on calibration, rather than accuracy.

The using of deep learning methods in medical images has been successfully used for various applications, including cell segmentation and deformability detection, thereby contributing significantly to advancements in medical analysis. Cell deformability is a fundamental criterion, which must be measured easily and accurately. One common approach for measuring cell deformability is to use microscopy techniques. Recent works have been efforts to develop more advanced and automated methods for measuring cell deformability based on microscopic images, but cell membrane segmentation techniques are still difficult to achieve with precision because of the quality of images. In this paper, we introduce a novel algorithm for cell segmentation that addresses the challenge of microscopic images. AD-MSC cells were controlled by a microfluidic-based system and cell images were acquired by an ultra-fast camera with variable frequency connected to a powerful computer to collect data. The proposed algorithm has a combination of two main components: denoising images using unsupervised learning and cell segmentation and deformability detection using supervised learning which aim to enhance image quality without the need for expensive materials and expert intervention and segment cell deformability with more precision. The contribution of this paper is the combination of two neural networks that treat the database more easily and without the presence of experts. This approach is used to have faster results with high performance according to low datasets from microscopy even with noisy microscopic images. The precision increases to 81 % when we combine DAE with U-Net, compared to 78 % when adding VAE to U-Net and 59 % when using only U-Net.

Integral to the success of transistor advancements is the accurate use of failure analysis (FA) which benefits in fine-tuning and optimization of the fabrication processes. However, the chip makers face several FA challenges as device sizes, structure, and material complexities scale dramatically. To sustain manufacturability, one can accelerate defect identification at all steps of the chip processing and design. On the other hand, as technologies scale below the nanometer nodes, devices are more sensitive to unavoidable process-induced variability. Therefore, metallic defects and process-induced variability need to be treated concurrently in the context of chip scaling, while failure diagnostic methods to decouple the effects should be developed. Indeed, the locating a defective component from thousands of circuits in a microchip in the presence of variability is a tedious task. This work shows how the SPICE circuit simulations coupled with machine learning based-physical modeling should be effectively used to tackle such a problem for a 6T-SRAM bit cell. An automatic bridge defect recognition system for such a circuit is devised by training a predictive model on simulation data. For feature descriptors of the model, the symmetry of the circuit and a fundamental material property are leveraged: metals (semiconductors) have a positive (negative) temperature coefficient of resistance up to a certain voltage range. Then, this work successfully demonstrates that how a defective circuit is identified along with its defective component's position with approximately 99.5 % accuracy. This proposed solution should greatly help to accelerate the production process of the integrated circuits.

In the context of survival analysis, data-driven neural network-based methods have been developed to model complex covariate effects. While these methods may provide better predictive performance than regression-based approaches, not all can model time-varying interactions and complex baseline hazards. To address this, we propose Case-Base Neural Networks (CBNNs) as a new approach that combines the case-base sampling framework with flexible neural network architectures. Using a novel sampling scheme and data augmentation to naturally account for censoring, we construct a feed-forward neural network that includes time as an input. CBNNs predict the probability of an event occurring at a given moment to estimate the full hazard function. We compare the performance of CBNNs to regression and neural network-based survival methods in a simulation and three case studies using two time-dependent metrics. First, we examine performance on a simulation involving a complex baseline hazard and time-varying interactions to assess all methods, with CBNN outperforming competitors. Then, we apply all methods to three real data applications, with CBNNs outperforming the competing models in two studies and showing similar performance in the third. Our results highlight the benefit of combining case-base sampling with deep learning to provide a simple and flexible framework for data-driven modeling of single event survival outcomes that estimates time-varying effects and a complex baseline hazard by design. An R package is available at https://github.com/Jesse-Islam/cbnn.

Generative Pre-trained Transformers (GPT) and Large language models (LLMs) have made significant advancements in natural language processing in recent years. The practical applications of LLMs are undeniable, rendering moot any debate about their impending influence. The power of LLMs has made them similar to machine learning models for decision-making problems. In this paper, we focus on binary classification which is a common use of ML models, particularly in credit lending applications. We show how a GPT model can perform almost as accurately as a classical logistic machine learning model but with a much lower number of sample observations. In particular, we show how, in the context of credit lending, LLMs can be improved and reach performances similar to classical logistic regression models using only a small set of examples.