It has taken 10 years, but Elon Musk has finally got to the punchline. The Tesla CEO has revealed the company's new car: the Model Y, the last part of one of Mr Musk's many long term plans. It means that the company now makes the Model S, Model 3, Model X and Model Y. Parked next to each other, the model numbers spell out S3XY. We'll tell you what's true. You can form your own view.

To evaluate the performance, agreement, and efficiency of a fully convolutional network (FCN) for liver lesion detection and segmentation at CT examinations in patients with colorectal liver metastases (CLMs). This retrospective study evaluated an automated method using an FCN that was trained, validated, and tested with 115, 15, and 26 contrast material–enhanced CT examinations containing 261, 22, and 105 lesions, respectively. Manual detection and segmentation by a radiologist was the reference standard. Performance of fully automated and user-corrected segmentations was compared with that of manual segmentations. The interuser agreement and interaction time of manual and user-corrected segmentations were assessed. Analyses included sensitivity and positive predictive value of detection, segmentation accuracy, Cohen κ, Bland-Altman analyses, and analysis of variance. Automated detection and segmentation of CLM by using deep learning with convolutional neural networks, when manually corrected, improved efficiency but did not substantially change agreement on volumetric measurements. Supplemental material is available for this article. A deep learning method shows promise for facilitating detection and segmentation of colorectal liver metastases; user correction of three-dimensional automated segmentations can generally resolve deficiencies of fully automated segmentation for small metastases and is faster than manual three-dimensional segmentation. Per-lesion sensitivity for lesions smaller than 10 mm was very low with automated segmentation (0.10) but was higher for user-corrected segmentation (0.30–0.57) and manual segmentation (0.58–0.70).

More and more frequently, schools across the United States are turning to artificial intelligence-backed solutions to stop tragic acts of student violence. Companies like Bark Technologies, Gaggle.net, and Securly, Inc., are using a combination of artificial intelligence (AI) and machine learning (ML) along with trained human safety experts to scan student emails, texts, documents, and in some cases, social media activity. They're looking for warning signs of cyber bullying, sexting, drug and alcohol use, depression, and to flag students who may pose a violent risk not only to themselves, but to classmates as well. Any potential problems discovered trigger alerts to school administration, parents, and law enforcement officials, depending on the severity. Bark ran a test pilot of its program with 25 schools in fall 2017. Bark chief parent officer, Titania Jordan, says, "We found some pretty alarming issues, including a bombing and school shooting threat."

We're into our second year of publishing a "Global AI Race" series of articles on artificial intelligence startups from around the world and it continues to pose a challenge. We use an objective measure of "total funding taken in so far" and that excludes any firms that choose not to disclose funding or are bootstrapped. We search for various categorizations like "artificial intelligence" or "deep learning" and that means we'll miss any firms that haven't chosen those categories in their Crunchbase profile. But the ones we worry about the most are those firms that we might include in one of our "top AI startups" lists that don't actually do AI. It's a huge problem, and one that was highlighted recently by a European venture capital firm, MMC Ventures, that surveyed 2,830 startups in Europe that were classified as being AI companies and found out that 44% of these companies were incorrectly classified as being "AI startups."

Quantum computing and artificial intelligence are both hyped ridiculously. But it seems a combination of the two may indeed combine to open up new possibilities. In a research paper published today in the journal Nature, researchers from IBM and MIT show how an IBM quantum computer can accelerate a specific type of machine-learning task called feature matching. The team says that future quantum computers should allow machine learning to hit new levels of complexity. As first imagined decades ago, quantum computers were seen as a different way to compute information.

NIMS, the University of Tokyo, Niigata University and RIKEN have jointly designed a multilayered metamaterial that realizes ultra-narrowband wavelength-selective thermal emission by combining the machine learning (Bayesian optimization) and thermal emission properties calculations (electromagnetic calculation). The joint team then experimentally fabricated the designed metamaterial and verified the performance. These results may facilitate the development of highly efficient energy devices. Thermal radiation, a phenomenon that an object emits heat as electromagnetic waves, is potentially applicable to a variety of energy devices, such as wavelength-selective heaters, infrared sensors and thermophotovoltaic generators. Highly efficient thermal emitters need to exhibit emission spectrum with narrow bands in practically usable wavelength range..

At first, Vaisagh Viswanathan, CTO at impress.ai, will talk about the challenges of hiring a tech team that works in a different country and timezone. And how he manages to use AI to help him scale the team while making sure major product milestones are met. Thereafter, Gaurang Torvekar, Co-founder and CEO of Indorse, a Skills Assessment Platform, will speak about his perspectives on hiring the right tech talent. He will also be sharing his experiences on how Indorse scaled its tech team beyond Singapore and all around the world. We want to dedicate more time to discussion and sharing among the community.

What did you think of the last commercial you watched? Would you buy the product? You might not remember or know for certain how you felt, but increasingly, machines do. New artificial intelligence technologies are learning and recognizing human emotions, and using that knowledge to improve everything from marketing campaigns to health care. These technologies are referred to as "emotion AI." Emotion AI is a subset of artificial intelligence (the broad term for machines replicating the way humans think) that measures, understands, simulates, and reacts to human emotions.

The 11th meeting of Forum for Information Retrieval Evaluation 2019 will be held in Kolkata, India. Started in 2008 with the aim of building a South Asian counterpart for TREC, CLEF and NTCIR, FIRE has since evolved continuously to meet the new challenges in multilingual information access. It has expanded to include new domains like plagiarism detection, legal information access, mixed script information retrieval and spoken document retrieval to name a few. Continuing the trend started in 2015, the FIRE will consist of a peer-reviewed conference track along with evaluation tasks. We invite full and short papers from information retrieval, natural language processing, and related domains.

Talk Abstract: In spite of great success of deep learning a question remains to what extent the computational properties of deep neural networks (DNNs) are similar to those of the human brain. The particularly non-biological aspect of deep learning is the supervised training process with the backpropagation algorithm, which requires massive amounts of labeled data, and a non-local learning rule for changing the synapse strengths. In this talk I will describes a learning algorithm that does not suffer from these two problems. It learns the weights of the lower layer of neural networks in a completely unsupervised fashion. The entire algorithm utilizes local learning rules which have conceptual biological plausibility.