Inspired by the recent success of deep learning in multiscale information encoding, we introduce a variational autoencoder (VAE) based semi-supervised method for detection of faulty traffic data, which is cast as a classification problem. Continuous wavelet transform (CWT) is applied to the time series of traffic volume data to obtain rich features embodied in time-frequency representation, followed by a twin of VAE models to separately encode normal data and faulty data. The resulting multiscale dual encodings are concatenated and fed to an attention-based classifier, consisting of a self-attention module and a multilayer perceptron. For comparison, the proposed architecture is evaluated against five different encoding schemes, including (1) VAE with only normal data encoding, (2) VAE with only faulty data encoding, (3) VAE with both normal and faulty data encodings, but without attention module in the classifier, (4) siamese encoding, and (5) cross-vision transformer (CViT) encoding. The first four encoding schemes adopted the same convolutional neural network (CNN) architecture while the fifth encoding scheme follows the transformer architecture of CViT. Our experiments show that the proposed architecture with the dual encoding scheme, coupled with attention module, outperforms other encoding schemes and results in classification accuracy of 96.4%, precision of 95.5%, and recall of 97.7%.

Classification using supervised learning requires annotating a large amount of classes-balanced data for model training and testing. This has practically limited the scope of applications with supervised learning, in particular deep learning. To address the issues associated with limited and imbalanced data, this paper introduces a sample-efficient co-supervised learning paradigm (SEC-CGAN), in which a conditional generative adversarial network (CGAN) is trained alongside the classifier and supplements semantics-conditioned, confidence-aware synthesized examples to the annotated data during the training process. In this setting, the CGAN not only serves as a co-supervisor but also provides complementary quality examples to aid the classifier training in an end-to-end fashion. Experiments demonstrate that the proposed SEC-CGAN outperforms the external classifier GAN (EC-GAN) and a baseline ResNet-18 classifier. For the comparison, all classifiers in above methods adopt the ResNet-18 architecture as the backbone. Particularly, for the Street View House Numbers dataset, using the 5% of training data, a test accuracy of 90.26% is achieved by SEC-CGAN as opposed to 88.59% by EC-GAN and 87.17% by the baseline classifier; for the highway image dataset, using the 10% of training data, a test accuracy of 98.27% is achieved by SEC-CGAN, compared to 97.84% by EC-GAN and 95.52% by the baseline classifier.

This short paper compiles the big ideas behind some philosophical views, definitions, and examples of causality. This collection spans the realms of the four commonly adopted approaches to causality: Hume's regularity, counterfactual, manipulation, and mechanisms. This short review is motivated by presenting simplified views and definitions and then supplements them with examples from various fields, including economics, education, medicine, politics, physics, and engineering. It is the hope that this short review comes in handy for new and interested readers with little knowledge of causality and causal inference. Introduction Causality is the science of cause and effect [1]. As identifying causal mechanisms is often regarded as a fundamental purist in most sciences, causality becomes elemental in advancing our knowledge. While causality is more profound in some research areas, the concept of causality is often vague or forgotten in others [2]. With the advent of data science and the ready accessibility of big data, there is a rising potential to leverage such data in pursuit of unlocking previously unknown, hidden mechanisms or perhaps confirming ongoing hypotheses and empirical knowledge [3]. Traditionally, researchers would collect data pertaining to a phenomenon and then analyze this data to describe it, creating a model that could be used to predict such a phenomenon and/or causally infer (or explain/understand) interesting questions about such a phenomenon (see Table 1) [4]. When the primary goal is to describe the data on hand, the researcher simply aims to visualize the data to tell its story.

Congratulations, you made it through another rough year. We've traditionally reserved the end of the year as an opportunity to have a little fun while shining a light on the year's biggest failings in tech, with a roundup affectionately known as CNET's annual Tech Turkeys. In the past, the ribbing was good natured, pointing out silly products or a random faux pas at a conference (hello, Michael Bay!). Then the problems across Big Tech piled on. Congressional hearings, election-swinging misinformation and privacy-invading breaches became a regular thing.

Retired four-star Gen. Jack Keane on President Zelenskyy's trip to D.C. and the need to continue supporting Ukraine. Ukrainian Foreign Minister Dmytro Kuleba said Monday that his country wants to hold a peace summit at the United Nations by the end of February but that Russia should first face a war crimes tribunal in an international court. "Every war ends in a diplomatic way. Every war ends as a result of the actions taken on the battlefield and at the negotiating table," Kuleba told the Associated Press on Monday. "The United Nations could be the best venue for holding this summit because this is not about making a favor to a certain country."

Tesla won't be able to advertise its cars as Full Self-Driving starting next year under a new California law. Senate Bill 1398 was signed into law by Gov. Gavin Newsome and it targets the electric automaker's marketing of software features in some models that imply the cars can fully drive themselves - which they cannot. Elon Musk's company lobbied against the bill, arguing that it already makes customers aware of the technology's limitations. Tesla owners in the U.S. must pay extra $15,000 for full self-driving software and were required to have a safety score of at least 80. However, Musk recently announced that anyone in North America can now request the FSD software. Tesla won't be able to advertise its cars as Full Self-Driving starting next year under a new California law Drivers are also told that regardless of the software's capabilities, they should always pay attention to the road and be prepared to take the wheel at any moment.

French authorities plan to use an AI-assisted crowd control system to monitor people during the 2024 Paris Olympics, according to a draft law seen by AFP on Thursday. The system is intended to allow the security services to detect disturbances and potential problems more easily, but will not use facial recognition technology, the bill says. The technology could be particularly useful during the highly ambitious open-air opening ceremony with Olympians sailing down the river Seine in front of a crowd of 600,000 people. French police and sports authorities faced severe criticism in May after shambolic scenes during the Champions League final in Paris when football fans were caught in a crowd crush and teargassed. The draft law, which was presented to the cabinet on Thursday, proposes other security measures including the use of full-body scanners and increases the sentences for hooliganism.

Artificial intelligence is evolving rapidly, with projects like OpenAI's DALL-E 2, Google's MINERVA, and DeepMind's Gato all pushing new technological boundaries. Until now, national governments have been slow to adopt this cutting-edge technology. In 2023, however, the opportunities to provide effective, targeted, and affordable services to citizens will prompt them to finally embrace AI, making government more transparent, accessible and effective. In some countries, AI is already being used to improve people's interaction with the state. This year, the Estonian government launched a new AI-based virtual assistant called Bürokratt.

Around 150 delegates from 50 participating countries took part in the recent plenary meeting of the ISO and IEC joint committee on artificial intelligence (ISO/IEC JTC 1/SC 42). At the meeting, delegates heard from the European Commission (EC) and approved a number of resolutions. The keynote speaker was Salvatore Scalzo, an EC Policy and Legal Officer in the field of AI. He works in the Directorate‑General for Communications Networks, Content and Technology, which develops and implements digital policies for Europe. Mr. Scalzo said that the EC was taking a strong interest in ISO/IEC AI standards as work continued on a future AI act.

On a hot summer's morning in July, Robert Oppenheimer stood in a control bunker in New Mexico and watched the results of his Manhattan Project burn the desert sand, transforming it into a mild but lightly radioactive green glass. Years later, when asked what went through his head when he saw that great grey cloud rise out of the sand, he said he was reminded of Hindu Scripture, the line from Vishnu: 'Now I am become Death, the destroyer of worlds'. Although, according to his brother, what he actually said after seeing the bomb explode was: 'I guess it worked'. As romantic as the potential of science can be, there is also a banality to the discoveries and inventions that shape our world. It is irrefutable that the atomic bomb changed the trajectory of the 20th century, ending the Second World War and fuelling the Cold War between the Soviet Union and the United States, and their proxies.