In many patients with diabetes, early signs leading to diagnosis include abnormal levels of glucose, increased urinary excretion, and higher food consumption. While those symptoms may be able to determine likelihood of diabetes diagnosis, machine learning could be the most capable way of accurately predicting future onset of diabetes, a new study has found. As published in the Journal of the Endocrine Society, a research team utilized a form of artificial intelligence known as machine learning, which comprises of computerized algorithms learning and adapting to new patterns when exposed to fresh data. For the study, over 500,000 medical records from more than 130,000 patients were analyzed by the research team, 65,505 of which were regarded as having no prior history of diabetes. The records, dated 2008 through 2018, were based in a populated region of Japan.

Much has been written about the role artificial intelligence (AI) can play in the fight against the Covid-19 pandemic. While the practicality of new technology is still being explored, companies are earnestly working on AI-enabled tools that could support the measures being taken to curb the outbreak. This includes tools to enforce social distancing, assess contact tracing, share valuable insights, as well as automated rapid tests, risk assessment apps, and resources to accelerate associated drug discovery initiatives, among others. While the widespread adoption of these AI-enabled solutions for these applications could take time due to appropriate scepticism around it, there is no contesting the fact that it can expedite efforts to manage the ongoing Covid-19 crisis. But while caution should still be exercised when it comes to experimental technologies – especially during a health emergency like this – there are less-explored areas amid this crisis where technology can add a lot of value without the accompanied risk. Incorporating AI into the mundane – but critical – administrative functions in a hospital can relieve these institutes of unprecedented strain on resources, especially doctors, physicians, and other medical practitioners.

The way to set your children up for success in this day and age is to ensure they learn about artificial intelligence, according to the billionaire tech entrepreneur Mark Cuban. "Give your kids an edge, have them sign up [and] learn the basics of Artificial Intelligence," Cuban tweeted on Monday. Cuban, who is a star on the hit ABC show "Shark Tank" and the owner of the Dallas Mavericks NBA basketball team, was promoting a free, one-hour virtual class his foundation is teaching an introduction to artificial intelligence in collaboration with A.I. For Anyone, a nonprofit organization that aims to improve literacy of artificial understanding. "Parents, want your kids to learn about artificial intelligence while you're stuck in quarantine," Cuban says on his LinkedIn account.

People around the world enjoy "virtual human" characters, whether in Hollywood films, Japanese anime, or video games. In recent years, AI-powered virtual humans have increasingly insinuated themselves into our daily lives. The virtual pop icon Teresa Teng has performed songs with Taiwanese singer Jay Chou, achieving huge success. The popular Chinese debate show "I CAN I BB" hosted a spirited episode on whether "Falling in love with an AI human can be considered true love or not," where many people argued it is possible for a human to fall in love with an AI. Are there limits to such human-machine relationships?

National Oceanic and Atmospheric Administration researchers, with help from engineers at Google, trained a neural network to distinguish humpback whale songs from other ocean noise. Researchers at the National Oceanic and Atmospheric Administration, with help from engineers at Google, trained a neural network to distinguish humpback whale songs from other ocean noise. The researchers used the resulting program to determine the occurrence of the species in islands in the Pacific, and how that may have changed over the past decade. Meanwhile, researchers at the Charles Stark Draper Laboratory and the New England Aquarium are using data from satellites, sonar, radar, human sightings, ocean currents, and more to train a machine learning algorithm to create a probability model to help locate the endangered North Atlantic right whale. Said the Monterey Bay Research Aquarium Institute's Kakani Katija, "What I love about technology or the progress we're seeing in AI, I think it's a hopeful time because if we get this right, I think it will have profound effects on how we observe our environment and create a sustainable future."

Artificial Intelligence (AI) is clearly a must-have when it comes to being competitive in today's markets. But implementing this technology has been challenging, even for some of the world's top companies. There are issues with data, finding the right talent and creating models that generate sufficient ROI. As a result, many AI projects fail. According to IDC, only abut 35% of organizations succeed in getting models into production successfully.

We consider the exploration-exploitation dilemma in finite-horizon reinforcement learning problems whose state-action space is endowed with a metric. We introduce Kernel-UCBVI, a model-based optimistic algorithm that leverages the smoothness of the MDP and a non-parametric kernel estimator of the rewards and transitions to efficiently balance exploration and exploitation. Unlike existing approaches with regret guarantees, it does not use any kind of partitioning of the state-action space. For problems with $K$ episodes and horizon $H$, we provide a regret bound of $O\left( H^3 K^{\max\left(\frac{1}{2}, \frac{2d}{2d+1}\right)}\right)$, where $d$ is the covering dimension of the joint state-action space. We empirically validate Kernel-UCBVI on discrete and continuous MDPs.

The treatment of malaria is a global health challenge that stands to benefit from the widespread introduction of a vaccine for the disease. A method has been developed to create a live organism vaccine using the sporozoites (SPZ) of the parasite Plasmodium falciparum (Pf), which are concentrated in the salivary glands of infected mosquitoes. Current manual dissection methods to obtain these PfSPZ are not optimally efficient for large-scale vaccine production. We propose an improved dissection procedure and a mechanical fixture that increases the rate of mosquito dissection and helps to deskill this stage of the production process. We further demonstrate the automation of a key step in this production process, the picking and placing of mosquitoes from a staging apparatus into a dissection assembly. This unit test of a robotic mosquito pick-and-place system is performed using a custom-designed micro-gripper attached to a four degree of freedom (4-DOF) robot under the guidance of a computer vision system. Mosquitoes are autonomously grasped and pulled to a pair of notched dissection blades to remove the head of the mosquito, allowing access to the salivary glands. Placement into these blades is adapted based on output from computer vision to accommodate for the unique anatomy and orientation of each grasped mosquito. In this pilot test of the system on 50 mosquitoes, we demonstrate a 100% grasping accuracy and a 90% accuracy in placing the mosquito with its neck within the blade notches such that the head can be removed. This is a promising result for this difficult and non-standard pick-and-place task.

Despite a multitude of empirical studies, little consensus exists on whether neural networks are able to generalise compositionally, a controversy that, in part, stems from a lack of agreement about what it means for a neural model to be compositional. As a response to this controversy, we present a set of tests that provide a bridge between, on the one hand, the vast amount of linguistic and philosophical theory about compositionality of language and, on the other, the successful neural models of language. We collect different interpretations of compositionality and translate them into five theoretically grounded tests for models that are formulated on a task-independent level. In particular, we provide tests to investigate (i) if models systematically recombine known parts and rules (ii) if models can extend their predictions beyond the length they have seen in the training data (iii) if models' composition operations are local or global (iv) if models' predictions are robust to synonym substitutions and (v) if models favour rules or exceptions during training. To demonstrate the usefulness of this evaluation paradigm, we instantiate these five tests on a highly compositional data set which we dub PCFG SET and apply the resulting tests to three popular sequence-to-sequence models: a recurrent, a convolution-based and a transformer model. We provide an in-depth analysis of the results, which uncover the strengths and weaknesses of these three architectures and point to potential areas of improvement.

Much progress has been made recently on text classification with methods based on neural networks. In particular, models using attention mechanism such as BERT have shown to have the capability of capturing the contextual information within a sentence or document. However, their ability of capturing the global information about the vocabulary of a language is more limited. This latter is the strength of Graph Convolutional Networks (GCN). In this paper, we propose VGCN-BERT model which combines the capability of BERT with a Vocabulary Graph Convolutional Network (VGCN). Local information and global information interact through different layers of BERT, allowing them to influence mutually and to build together a final representation for classification. In our experiments on several text classification datasets, our approach outperforms BERT and GCN alone, and achieve higher effectiveness than that reported in previous studies.