Depending on who you ask, artificial intelligence is either going to save humanity or kill us all. At the same time, AI seems to be in everything from home assistants to washing machines. But what is AI, and why is everyone going on about it? Here's how to explain artificial intelligence to your family over Christmas when they ask whether they've unwittingly inviting the AI apocalypse into their living room. Roughly speaking, artificial intelligence research is all about creating computers that can perform tasks that usually require human intelligence.

Want to watch this again later? Sign in to add this video to a playlist. Report Need to report the video? Sign in to report inappropriate content. Report Need to report the video?

Google is one company, like Microsoft, that has been spearheading artificial intelligence (AI) development. The two tech giants believe AI will and indeed is revolutionizing our lives. However, there are plenty of bumps on the road. Chinese social and gaming company Tencent says it has found a "significant security loophole" in Google's machine-learning platform TensorFlow. Tencent is a social gaming giant that is one of Android's leading developers.

Megvii's Co-founder and CEO Yin Qi said, "From an academic research team to a commercialized company, I have always sought to balance technical faith and commercial value. The deeper we go into the industry, the more we realize how important scientific research it is, so this year, we set up an academic committee and invited renowned scholars like Andrew Yao Chi-Chih and Nanning Zheng as our consultants. Meanwhile we also hope that we can establish relations with first-class universities such as HKUST, so then, we built the joint laboratory. I think today is a brand new start to both sides, we will regard this as a starting point to our future success in scientific and technological cooperation between HKUST and Megvii."

We propose a stochastic variance reduced optimization algorithm for solving sparse learning problems with cardinality constraints. Sufficient conditions are provided, under which the proposed algorithm enjoys strong linear convergence guarantees and optimal estimation accuracy in high dimensions. We further extend the proposed algorithm to an asynchronous parallel variant with a near linear speedup. Numerical experiments demonstrate the efficiency of our algorithm in terms of both parameter estimation and computational performance.

The estimation of the conditional mode, or modal regression [24, 11, 5, 22], is an important topic in statistics [21, 25, 24], econometrics [16, 17, 8, 15, 11], and machine learning [7, 22]. Compared to ordinary regression, modal regression is particularly useful when the data distribution is highly skewed and has fat tails. 1 In such a situation, ordinary regression, which estimates the conditional mean of the distribution, fails to capture the major trend underlying the data. This is because the conditional mean is not necessarily the point where the data points distribute densely, i.e., it can be far away from the majority of the data. Conditional mode is a convenient alternative to the conditional mean in this situation as it can capture the majority of the data. Hence, with modal regression, we can find a major trend underlying the data.

We develop efficient algorithms for estimating low-degree moments of unknown distributions in the presence of adversarial outliers. The guarantees of our algorithms improve in many cases significantly over the best previous ones, obtained in recent works of Diakonikolas et al, Lai et al, and Charikar et al. We also show that the guarantees of our algorithms match information-theoretic lower-bounds for the class of distributions we consider. These improved guarantees allow us to give improved algorithms for independent component analysis and learning mixtures of Gaussians in the presence of outliers. Our algorithms are based on a standard sum-of-squares relaxation of the following conceptually-simple optimization problem: Among all distributions whose moments are bounded in the same way as for the unknown distribution, find the one that is closest in statistical distance to the empirical distribution of the adversarially-corrupted sample.

Adding to the list of things robots do better than humans, this creation from the University of Tokyo is now better than we are at exercise. In Science Robotics today, the researchers showed off their newest humanoid robot, Kengoro, who can do sit-ups, push-ups, and back extensions -- which a quick Google search revealed to me, is definitely a real thing. Kengoro is nothing without a series of 116 actuators. The actuators are placed inside its frame in an attempt to mirror the human body, and through push/pull commands can contract much like whatever flab resides where our muscles should be. These actuators, though, produce heat.

An Israeli man is the first patient ever to receive a surgical procedure Tuesday to regrow part of his bone, according to reports. After a car accident eight months prior, the male patient had part of his shin bone removed. Surgeons were able to successfully tend to his injury with a procedure that's being deemed as "science fiction." The operation was developed several years back by Bonus BioGroup, an Israel-based biotechnology company that utilizes technology to generate tissue-regenerating bone grafts. The company's medical team harvested fat tissue cells from the man's leg and grew them in their lab. The fat cells were then injected back into the man's leg to regenerate the bone's missing parts.