Artificial Intelligence ("AI") is no longer the stuff of sci-fi movies or alien invasions. The technology has permeated everyday life from Siri and Alexa to Facebook and Google. While marketing teams have been relying on AI for years to help streamline business efforts and target consumers, employers have finally joined in on the hype. While the use of AI can be an efficient and cost effective means for employers to handle tasks such as talent acquisition, compensation analysis, and administrative functions, it is not without its challenges. As lawmakers on the federal and state level struggle to catch up with the rapidly changing technology, it is imperative for employers to stay ahead of the curve and ensure that their use of AI is not exposing them to costly litigation. AI is often used in the workplace to assist employers with recruitment through the use of algorithms to make hiring decisions.

The promise of ar ti fi cial in tel li gence is that ma chines can wean man u fac tur ers away from the breadth of re sources it takes to dis cov er a po ten tial ly po tent com pound, and lever age the med ical con di tions it could be used to treat. Alex Zha voronkov, whose Hong Kong-based AI-shop In sil i co Med i cine has en trenched it self in the ten ta cles of bio phar ma R&D, now has some da ta to bol ster the fer vor and in vest ment that AI has drummed up. On Mon day, Zha voronkov and his fel low AI scouts from Wuxi and sci en tists from the Uni ver si ty of Toron to, pub lished a pa per in Na ture Biotech nol o gy that sup port ed the fea si bil i ty of em ploy ing a ma chine learn ing ap proach -- gen er a tive ad ver sar i al net works (GANs) and re in force ment learn ing -- for de no vo drug de sign. The re searchers found that they were able to iden ti fy a myr i ad of com pounds tar get ing a pro tein called Dis coidin do main re cep tor 1 (DDR1) -- which is ex pressed in ep ithe lial cells and in volved in fi bro sis -- in a swift 21 days. Af ter that, six mol e cules were then se lect ed for syn the sis in the lab -- those tests re vealed four with po ten tial.

The university installs face recognition system in the classroom.

Technologies such as artificial intelligence (AI) and virtual reality (VR) are rapidly expanding opportunities for teaching and learning, and they are giving college administrators new and different ways to track student outcomes. To learn more about the impact of these technologies, we attended a handful of panels on the topic led by higher education and technology leaders at Educause's annual conference in Denver this week. From teaching with VR to tracking student success with AI, we explore how colleges and universities are using new technologies to conduct research, teach students and create smarter campuses. Virtual and augmented reality tools can provide students with experiences that would be otherwise too expensive or even impossible to replicate in the real world, from exploring the inside of a cell to traversing faraway planets, said D. Christopher Brooks, director of research at the Educause Center for Analysis and Research. At Hamilton College, for example, these tools are changing the way the 1,850-student liberal arts institution teaches human anatomy.

The Danish toy company has launched a global trial with four building sets and hopes to release more next year depending on user feedback. Using the design script used by Lego creators, known as LXFML data, the Austrian Research Institute for Artificial Intelligence developed software that is able to turn the instructions into braille and English audio. The idea came about after blind entrepreneur and Lego enthusiast Matthew Shifrin previously relied on a friend to translate instructions into braille, which enabled him to build independently for the first time. After his friend died, Mr Shifrin was introduced to Lego's Creative Play Lab and decided to push for development of the technology: "This is extremely important for blind children because there aren't a lot of places where we can say'Look Mum and Dad! I built this on my own... I did this'," he said.

Fake news and social media posts are such a threat to U.S. security that the Defense Department is launching a project to repel "large-scale, automated disinformation attacks," as the top Republican in Congress blocks efforts to protect the integrity of elections. The Defense Advanced Research Projects Agency wants custom software that can unearth fakes hidden among more than 500,000 stories, photos, videos and audio clips. If successful, the system after four years of trials may expand to detect malicious intent and prevent viral fake news from polarizing society. "A decade ago, today's state of the art would have registered as sci-fi -- that's how fast the improvements have come," said Andrew Grotto at the Center for International Security at Stanford University. "There is no reason to think the pace of innovation will slow any time soon."

We provide the first polynomial-time algorithm for recognizing if a profile of (possibly weak) preference orders is top-monotonic. Top-monotonicity is a generalization of the notions of single-peakedness and single-crossingness, defined by Barbera and Moreno. Top-monotonic profiles always have weak Condorcet winners and satisfy a variant of the median voter theorem. Our algorithm proceeds by reducing the recognition problem to the SAT-2CNF problem.

JPEG2000 (j2k) is a highly popular format for image and video compression.With the rapidly growing applications of cloud based image classification, most existing j2k-compatible schemes would stream compressed color images from the source before reconstruction at the processing center as inputs to deep CNNs. We propose to remove the computationally costly reconstruction step by training a deep CNN image classifier using the CDF 9/7 Discrete Wavelet Transformed (DWT) coefficients directly extracted from j2k-compressed images. We demonstrate additional computation savings by utilizing shallower CNN to achieve classification of good accuracy in the DWT domain. Furthermore, we show that traditional augmentation transforms such as flipping/shifting are ineffective in the DWT domain and present different augmentation transformations to achieve more accurate classification without any additional cost. This way, faster and more accurate classification is possible for j2k encoded images without image reconstruction. Through experiments on CIFAR-10 and Tiny ImageNet data sets, we show that the performance of the proposed solution is consistent for image transmission over limited channel bandwidth.

Distribution mismatches between the data seen at training and at application time remain a major challenge in all application areas of machine learning. We study this problem in the context of machine listening (Task 1b of the DCASE 2019 Challenge). We propose a novel approach to learn domain-invariant classifiers in an end-to-end fashion by enforcing equal hidden layer representations for domain-parallel samples, i.e. time-aligned recordings from different recording devices. No classification labels are needed for our domain adaptation (DA) method, which makes the data collection process cheaper.

The multi-armed bandit problem has been extensively studie d in computer science, operations research and economics since the seminal work of Robb ins (1952). It is a model designed for sequential decision-making in which a player c hooses at each time step amongst a finite set of available arms and receives a reward for the cho sen decision. The player's objective is to minimize the difference, called regret, betwee n the rewards she receives and the rewards accumulated by the best arm. The rewards of each arm i s drawn from a probability distribution in the stochastic multi-armed bandit problem; but in adversarial multi-armed bandit models, there is typically no assumption imposed on t he sequence of rewards received by the player. In recent work, Lykouris et al. (2018) introduce a model in wh ich an adversary could corrupt the stochastic reward generated by an arm pull. They provide an algorithm and show that the regret of this "middle ground" scenario degrad es smoothly with the amount of corruption injected by the adversary. Gupta et al. (2019) pr esent an alternative algorithm which gives a significant improvement.