Singularity Pulse  
  Information Technology

IBM To Launch Drones To Deliver Coffee To Its Employees

IBM has recently filed an application for a patent on a system of camera-equipped drones that could detect pupil dilation and facial expressions. The drones fall into a category of biometric-sensing drones. As a consequence of fatigue, before workers start making typos and errors, these drones will be able to foresee the dizziness and provide that particular employee with coffee cups. The company is hopeful that this will increase their overall efficiency and effective man hours.


New hybrid AI Camera Makes Autonomous Car More Efficient

Stanford University research team has developed a technology to shrink down the computing power and sensors needed in an automated moving car. A new AI camera system will classify images faster, making it more energy efficient. Two computers are combined in one, creating a hybrid “optical-electrical computer” for image analysis. Its first layer physically pre-processes image data by naturally filtering light without a power source, thus saving lot of time and energy. Additionally, there are fewer calculations to be done as the math of AI has been outsourced to the optics of the camera.

  Space Research

NASA is Developing AI to Give Directions in Space

NASA Frontier Development Lab (FDL) recently concluded its eight week program which included nine teams working on core problems for space travel and extraterrestrial colonization. Its goal was to address these specific knowledge gaps using artificial intelligence. One of the topics of the program was the problem of extraterrestrial navigation for both humans and robots, and what solutions can address them with today’s deep learning techniques. The team at NASA plans to teach an AI what the moon looks like by feeding it millions of images and then use a neural network to create a virtual moon. This AI-powered solution should work for any surface we’re able to take images of – including Mars. It likely won’t have the same efficacy for open space localization but flying through space presents a different challenge than navigating an alien surface.


ML Powered Fake News Detection Algorithm Outperforms Humans

Researchers from University of Michigan and University of Amsterdam have developed an algorithm for detecting fake news. It uses NLP to search for specific patterns to determine if a particular article is fake news. Dataset was crowd sourced asking Amazon Mechanical Turk workers to reinterpret 500 real news stories as fake news stories. Participants imitated the journalist’s style but fudged the facts. Algorithm was fed both, fake and real news and it taught itself to make a distinction. Later on, both real and fake news were fed directly from web and it outperformed humans in making distinctions.

Social Media  
  Health Care

ML Algorithms To Help Humans Work Out Smartly

A company named Tonal is using ML techniques to make work outs more intense for humans. Tonal workout system is a wall mounted machine that lets its users select the type of workout they want to do or the muscle group they want to target. It then automatically plays a video of that request too. It uses electromagnetic engine that can replicate the resistance of weights. It monitors users’ progress and automatically adjust difficulty levels.


AI To Predict Musical Taste Genres Of People

Researchers at Jönköping University and Maastricht University have published a paper where they described a system that considers a person’s listening behaviors and, using machine learning algorithms and psychological models, infers their “musical sophistication.” The data is collected through an app which taps Spotify’s API. This allows retrieval of users’ playlists and audio features along with artist popularity. It was fed into a neural network. The algorithm predicted subjects’ emotions and active musical engagement with more than 90% accuracy.

  Online Gaming

Humans Still Better Than AI At Playing Dota 2

Elon Musk’s AI research company’s team, “Open AI Five” lost to human players in a five-versus-five match of the Dota 2 video game. The team lost first of its three matches being held in Vancouver. Company’s co-founder Greg Brockman had said that the system had “less than a 50 percent probability of winning.” The team trained itself on equivalent of 180 years of game hours, using a procedure called “Proximal Policy Optimization” powered by 128,000 processor cores and 256 graphics cards. Dota 2 has around 1000 valid moves in each moment and the system makes 7.5 observations per second of gameplay.


Sony Launches AI Powered Robot Dog

Sony has announced the launch of its Aibo robotics dogs by the year-end holiday season. The robot pup is infused with image sensors and AI, enabling it to recognize faces and predict behavior. The owner of the dog will have an application which can be used to adjust its settings, add or remove tricks and store memories on cloud. One will also be able to check on internet-linked companions while being away from their home, watching through their dog’s robotic eyes.


Datamaran To Have Patents For ESG Business & Investments

Datamaran, a SaaS provider for non-financial risk management, has announced publication of three of its patent applications. The three patents are available on the World Intellectual Property Organization website. The patents are for first, "A method and system for providing business intelligence", second, "A method and systems for locating regulatory information" and third, “"A method and systems for processing data using augmented natural language processing engine". These patents are expected to help company enable a systematic and thorough monitoring and analysis of Environmental Social Governance (ESG) risks.

“Quantum Computing, a revolution in technology, has shown great promise towards solving complex computing problems currently outside the capabilities of current computers. Though in its infancy, we at Decimal Point Analytics strongly believe that it is going to grow exponentially in the near future. It holds the potential not only to boost the AI revolution but also transform the way data is synthesized. The articles on Quantum computing will give an insight into the recent developments in this space.”
  Quantum Computing

IBM’s “Qiskit” To Provide Open Source Quantum Computing Framework

Qiskit (Quantum Information Science Kit) is IBM’s open source framework which provides access to real quantum computing in the cloud to everyone. Qiskit was launched a year ago. Researchers are making use of qubits to solve problems that were previously too difficult to solve on average computers. The interface is that of Python, for ease of expressing. Hobbyists and developers are leveraging technology made available on platform to write quantum programs, created games and compose music using real quantum computers.


Germany Produces Quantum Transistor From A Single Atom

Researchers at Karlsruhe Institute of Technology have developed a quantum transistor by using just a single atom. The developed transistor can even operate at room temperature. The new transistor moves a single atom of silver to open and close a circuit and is described as world’s smallest transistor. It is termed as “quantum” switch because it can perform even more complex information than the today’s conventional binary “on-off” signal of today’s transistors.

Quantum Computing  
Contact us:

5A, B-Wing, Trade Star Building, J. B. Nagar, Andheri-Kurla Road, Andheri (East), Mumbai - 400 059, Maharashtra, India
E:   T: +91 22300 15200