ArXiv March 13, 2018 Fractal AI is a theory for general artificial intelligence. It allows to derive new mathematical tools that constitute the foundations for a new kind of stochastic calculus. Among other things, Fractal AI makes it possible to generate a huge database of top performing examples with very little amount of computation required, transforming Reinforcement Learning into a supervised problem. The new techniques presented here have direct applications to other areas such as: Non-equilibrium thermodynamics, chemistry, quantum physics, economics, information theory, and non-linear control theory… read more. Open Access TECHNICAL ARTICLE
Category Archives: Artificial Intelligence
In five years quantum computing will be mainstream
Next Big Future March 19, 2018 IBM Researchers are already reaching major quantum chemistry milestones, having recently used a quantum computer to successfully simulate atomic bonding in beryllium hydride (BeH2), the most complex molecule ever simulated by a quantum computer. In the future quantum computers will continue to address problems with ever-greater complexity, eventually catching up to and surpassing what we can do with classical machines alone… read more.
The “Black Mirror” scenarios that are leading some experts to call for more secrecy on AI
MIT Technology Review February 21, 2018 A report, “The Malicious Use of Artificial Intelligence: Forecasting, Prevention, and Mitigation” authored by a team of more than 20 researchers in the UK, warns that AI creates new opportunities for criminals, political operatives, and oppressive governments—so much so that some AI research may need to be kept secret. The report includes four dystopian vignettes involving artificial intelligence to illustrate just a handful of the risks – the smarter phishing scam, the malware epidemic, the robot assassin, and a bigger Big Brother. It recommends more research and debate on the risks of AI and […]
New algorithm can create movies from just a few snippets of text
Science Magazine February 23, 2018 Researchers in Belgium have developed a machine learning a neural network algorithm. During training, software assesses its performance after each attempt, and feedback circulates through the millions of network connections to refine future computations. The first stage uses the text to create a “gist” of the video, the second stage takes both the gist and the text and produces a short video. During training, a second network acts as a “discriminator.” As it gets better, it becomes a harsher critic, and its feedback sets a higher bar for the generator network. Currently, the videos are […]
Chemists harness artificial intelligence to predict the future (of chemical reactions)
Eurekalert February 15, 2018 To manufacture medicines, chemists must find the right combinations of chemicals to make the necessary chemical structures. A team of researchers in the US (Princeton University, industry partner) has developed a software to accurately predict reaction yields while varying up to four reaction components. Reaction yields can be accurately predicted using the results of ‘only’ hundreds of reactions (instead of thousands). The code extracts descriptors for each chemical used in the model calculating quantitative descriptors for each chemical, to use as inputs for the model. The software can work for any reaction, any substrate. The idea […]
Why even a moth’s brain is smarter than an AI
MIT Technology Review February 19, 2018 Some critical machine-learning mechanisms have no analogue in the natural world, where learning seems to occur in a different way. Researchers at the University of Washington have created an artificial neural network that mimics the structure and behavior of the olfactory learning system in Manduca sexta moths. Their model can robustly learn new odors, and their simulations of integrate-and-fire neurons match the statistical features of in vivo firing rate data. This work that could have significant implications for the design of synthetic neural networks that need to learn quickly…read more. Open Access TECHNICAL ARTICLE
Researchers help robots think and plan in the abstract
Eurekalert February 8, 2018 A team of researchers in the US (Brown University, MIT) there has been less progress in perceptual abstraction, which has to do with helping a robot make sense of its pixelated surroundings. They gave a robot called Ana a set of high-level motor skills for manipulating the objects in a room. Once Ana was armed with her learned abstract representation, the researchers asked her to do something that required some planning which Ana executed. According to the researchers, their research provides an important theoretical building block for applying artificial intelligence to robotics… read more. Video Open […]
Artificial intelligence is the weapon of the next Cold War
Physorg January 29, 2018 During the Cold War, the weapon of choice was nuclear missiles. Today it is software, whether it is used for attacking computer systems or targets in the real world. AI may be the way Russia can rebalance the power shift created by the U.S. outspending Russia nearly 10-to-1 on defense each year. AI can also be used to control non-nuclear weapons including unmanned vehicles like drones and cyberweapons. Countries might agree to a proposed Digital Geneva Convention to limit AI conflict. But that won’t stop AI attacks by independent nationalist groups, militias, criminal organizations, terrorists and […]
Engineers design artificial synapse for “brain-on-a-chip” hardware
Kurzweil AI January 22, 2018 A team of researchers in the US (MIT, Arizona State University) has demonstrated analog resistive switching devices that possess desired characteristics for neuromorphic computing networks with minimal performance variations. They used a single-crystalline SiGe layer epitaxially grown on Si as a switching medium. It utilizes threading dislocations in SiGe to confine metal filaments in a defined, one-dimensional channel which results in drastically enhanced switching uniformity and long retention/high endurance with a high analog on/off ratio. In tests their neural network hardware recognized handwritten samples 95 percent of the time, compared to the 97 percent accuracy […]
Beijing Is Getting a $2.1 Billion AI District
MIT Technology Review January 4, 2018 China is gearing up to build a technology park in Beijing entirely dedicated to the development of artificial intelligence. The endeavor is just the latest sign of China’s remarkable ambition to master and dominate artificial intelligence by 2020. The plan will apparently establish a “national AI research center” and will include efforts to form partnerships with foreign research institutions and companies. China’s AI masterplan seems to have been inspired, in part, by the the U.S. government report at the very end of the Obama administration, PREPARING FOR THE FUTURE OF ARTIFICIAL INTELLIGENCE. Related document National […]