What’s happened in AI: October 8th-14th

By | October 17, 2018

Another week, another win for Google’s AI program. This time, its Lymp Node Assistant (LYNA) achieved 99% accuracy in metastatic breast cancer detection. Hopefully these types of applications of AI continue, in particular, to other types of cancer.

Meanwhile on the autonomous vehicle front, we have numerous developments from companies and governments. Arizona is launching an autonomous vehicle institute, self-driving taxis are now roaming in Dubai’s streets, Waymo’s vehicles hit the 10 million mile marker on public roads, Microsoft invested in Singapore’s grab, and much more.

Company developments:

Google AI claims 99% accuracy in metastatic breast cancer detection – Oct. 12, 2018 (VentureBeat)

  • Metastatic tumors — cancerous cells which break away from their tissue of origin, travel through the body through the circulatory or lymph systems, and form new tumors in other parts of the body — are notoriously difficult to detect. A 2009 study of 102 breast cancer patients at two Boston health centers found that one in four were affected by the “process of care” failures such as inadequate physical examinations and incomplete diagnostic tests
  • Google’s AI system — dubbed Lymph Node Assistant, or LYNA — is described in a paper titled “Artificial Intelligence-Based Breast Cancer Nodal Metastasis Detection,” published in The American Journal of Surgical Pathology. In tests, it achieved an area under the receiver operating characteristic (AUC) — a measure of detection accuracy — of 99 percent. That’s superior to human pathologists, who according to one recent assessment miss small metastases on individual slides as much as 62 percent of the time when under time constraints

Waymo’s driverless cars hit a new milestone: 10 million miles on public roads – Oct. 10, 2018 (The Verge)

  • As it nears the launch of its first commercial service, Waymo is celebrating a new milestone: 10 million miles driven on public roads. In recognition of this threshold, the Alphabet company released a video that features a lot of new footage of its cars operating without a human driver behind the wheel. In addition, Waymo’s top executive is setting realistic expectations about some of the limitations of its advanced technology

Volvo’s next generation of cars will use Nvidia’s self-driving car platform – Oct. 10, 2018 (The Verge)

  • Volvo has announced that it will use Nvidia’s Drive AGX Xavier computer for its next generation of vehicles. The hardware, which was announced by Nvidia in September, has the power to be able to handle full autonomy within controlled areas (so-called Level 4 autonomy), but it will launch with “Level 2+” capabilities, placing it on a similar level as current Tesla models

Huawei announces two AI chips as China continues its move away from US silicon – Oct. 10, 2018 (MIT News)

  • The Chinese telecommunications giant has announced two chips optimized for artificial intelligence. The chips reflects an important shift in China’s tech ambitions as the country seeks to lessen its dependence on foreign microprocessors. Chinese tech companies are among the most innovative and profitable in the world, but they rely heavily on chips made by US companies such as Intel, Qualcomm, and Nvidia. This is a troubling for those companies, and for the Chinese government, because microprocessors are a foundational technology that often enable innovation in other areas

Magic Leap’s Mica is a human-like AI in augmented reality – Oct. 10, 2018 (VentureBeat)

  • Interesting read on one writer’s take when Magic Leap showed off a demo of Mica, a humanlike artificial intelligence that can be viewed in the company’s augmented reality glasses, the Magic Leap One Creator Edition

Amazon reportedly scraps internal AI recruiting tool that was biased against women – Oct. 10, 2018 (The Verge)

  • Bias in machine learning can be a problem even for companies with plenty of experience with AI, like Amazon. According to a report from Reuters, the e-commerce giant scrapped an internal project that was trying to use AI to vet job applications after the software consistently downgraded female candidates
  • Because AI systems learn to make decisions by looking at historical data they often perpetuate existing biases. In this case, that bias was the male-dominated working environment of the tech world. According to Reuters, Amazon’s program penalized applicants who attended all-women’s colleges, as well as any resumes that contained the word “women’s” (as might appear in the phrase “women’s chess club”)

Instagram is using AI to detect bullying in photos and captions – Oct. 9, 2018 (The Verge)

  • The new feature will roll out to users in the coming weeks, launching in time for October’s National Bullying Prevention Month in the US and just before Anti-Bullying Week in the UK. The same technology is also being added to live videos to filter comments there as well
  • This is the first product announcement under new Instagram chief Adam Mosseri who took over following the hasty departure of co-founders Kevin Systrom and Mike Krieger last month. The split was reportedly due to simmering tensions between the pair and parent company Facebook, which has frequently meddled with Instagram’s product

Huawei commits to ‘huge investment’ in top AI talent with eye on the long-term benefit – Oct. 9, 2018 (South China Morning Post)

  • Huawei Technologies, the world’s largest telecommunications equipment supplier, said it will continue to invest heavily in artificial intelligence (AI) manpower, despite the added cost and pressure on the company’s bottom line, because the long-term payoff will be worthwhile
  • The adoption of AI technologies will not only make its products more intelligent and improve internal management efficiency, it could contribute as much as 90 per cent of total income even though AI headcount would be relatively small compared to the total number of employees, Huang Weiwei, a senior management consultant at Huawei, said on Tuesday

Ford’s patent lets you pilot a self-driving car with your phone – Oct. 8, 2018 (CNET)

  • CarBuzz alerted us to a Ford patent for “non-autonomous steering modes.” The patent offers a pretty straightforward method to turn a personal touchscreen device (such as a phone) into something capable of driving a vehicle. Yes, that’s right, all those years of Mario Kart could actually pay off
  • The patent describes multiple ways this could happen. First, the device could be used as an ersatz steering wheel. After the system detects a request to handoff the responsibility of driving from vehicle to person, it will sync the device to the vehicle, determining a “base” position for the wheel. After that, you can rotate the device to steer the vehicle, like you would in a mobile game.


Apple bought a machine learning green screen startup to focus on AR – Oct. 10, 2018 (The Verge)

  • Apple has quietly bought Spektral, a Danish machine learning startup that specializes in real-time green screen technology. The $30 million deal actually happened last year, but it was reported today by Danish newspaper Børsen. Apple has been focusing more and more on its AR capabilities lately, and this latest acquisition may be meant to boost the iPhone’s AR features for Memoji or FaceTime or as a part of its plans for an augmented reality headset, which Bloomberg reported may be coming in 2020
  • Spektral, which previously went by the name CloudCutout, uses machine learning and computer vision techniques to “cut out” people from video backgrounds in real time on smartphones. “Combining deep neural networks and spectral graph theory with the computing power of modern GPUs, our engine can process images and video from the camera in real-time (60 fps) directly on the device,” the company explained on its website

Fundraising / investment:

Micron’s Venture Fund Bets $100 Million on Artificial Intelligence – Oct. 13, 2018 (Electric Design)

  • Micron Technology announced a $100 million venture fund on Wednesday that will invest in artificial intelligence startups. The company is betting that broader use of the technology will increase sales of the company’s memory chips in data centers where the algorithms are trained and the systems where they are used to spot imperfections in products on a manufacturing line or to understand human speech
  • These trends are at the heart of the biggest opportunities in front of us, and increasingly require memory and storage technologies to turn vast amounts of data into insights,” said Sanjay Mehrotra, the company’s chief executive, in a statement. The neural networks at the heart of machine learning are growing more and more complicated. Handling them means using not only more compute but also more memory and storage

Deep Lens raises $3.2 million for AI cloud pathology platform – Oct. 12, 2018 (VentureBeat)

  • Deep Lens’ flagship product is Virtual Imaging for Pathology Education and Research, or VIPER. It’s a full stack solution consisting of an artificially intelligent (AI) image detection suite, feedback tools, cloud storage, and APIs for integration with hardware and applications — all designed to facilitate peer-to-peer collaboration and clinical research, Billiter said
  • VIPER has always been about empowering and enabling the pathology community, allowing them to focus on the nuanced diagnoses and case-specific details that require many years of specialized medical training,” he said. “It’s incredibly exciting to now leverage the evolution in machine vision technology as we launch VIPER globally, free of charge to benefit all pathologists and, in turn, their patients.”

June.ai looks to take on the inbox with $1.5 million – Oct. 11, 2018 (TechCrunch)

  • Co-founder and CEO Allie Sutton said that the team didn’t look at the way people use email as a starting point, but rather looked at the way people deal with communication and information flow on the whole. That then led June to build a platform that doesn’t necessarily replace email, but is rather backwards compatible with email services
  • With $1.5 million in funding, June isn’t immediately concerned with revenue but rather looking to spread the word about the platform and promote adoption. June works with a number of the biggest email providers, including Gmail, Yahoo!, and Outlook. Sutton said that adoption will be the biggest challenge. “Folks have built the way in which they process emails,” said Sutton. “Some over a few years and some over decades. We’ve built a more efficient way for them to process that information, but giving users enough explanation and information that they can adopt this new way of gathering and executing on information communication is our biggest challenge. We need to prove it’s more efficient for them.”

RBC Foundation supports advancing ethical AI with $1 million commitment to CIFAR – Oct. 9, 2018 (Newswire)

  • At the official opening of the Borealis AI Montréal lab today, Dave McKay, President & CEO of RBC, announced the RBC Foundation will donate $1 million over the next three years to the Canadian Institute for Advanced Research (CIFAR). The commitment will support research and initiatives aimed at furthering the study of ethical artificial intelligence (AI) practices
  • CIFAR is leading the , working in partnership with three world renowned AI institutes – the Alberta Machine Intelligence Institute (Amii) in Edmonton, Mila in Montreal and the Vector Institute in Toronto. CIFAR will work closely with , an RBC institute for research, to deliver on key aspects of the strategy

Microsoft invests in Grab to bring AI and big data to on-demand services – Oct. 8, 2018 (TechCrunch)

  • Microsoft has made a strategic investment in ride-hailing and on-demand services company Grab as part of a deal that includes collaborating on big data and AI projects. Under the agreement, Singapore-based Grab will adopt Microsoft Azure as its preferred cloud platformAzure cloud computing service. Microsoft and Grab didn’t disclose financial terms
  • The companies said they’ll also work on several “innovative deep technology projects,” including new authentication measures such as facial recognition with built-in AI (for drivers and customers who opt in) to replace the old-school method of checking IDs. The companies will also investigate uses for natural language processing, machine learning and AI in Grab’s platform such as map creation, fraud detection services and the ability for passengers to take a photo of their current location and have it translated into an actual address for the driver.

Machinify raises $10 million to help businesses use AI to monetize data – Oct. 8, 2018 (TechCrunch)

  • The artificial intelligence company just raised a $10 million Series A round led by Battery Ventures with participation from GV and Matrix Partners. “Our core notion is that today, enterprises are collecting a ton of data,” Machinify founder and CEO Prasanna Ganesan told TechCrunch. “But if you look at how many of them are successful in turning it into smarter decision-making to drive efficiency, very few companies are succeeding.”
  • With Machinify, enterprise customers feed the system raw data, specify what they’re trying to optimize for — whether that be revenue or some other goal — and then the machine figures out what to do from there. Based on past decisions, the machine can figure out the right thing to do, Ganesan said

Chinese AI Firm CloudWalk Raises Series B+ Round To Bring Total Financing To $509M – Oct. 8, 2018 (China Money Network)

  • Investors in this series B+ financing round are from state-owned investment institutions, including China Reform Fund, Atlas Capital Group, SFund International Holdings Limited, Technology Financial Group and Bohai Industrial Investment Fund, said the company in a statement released today in its WeChat official account
  • CloudWalk Technology plans to use the new funding in big data-related researches, development and expansion of the country’s AI application platforms, construction of research centers, talent cultivation plans, and the company’s industrial layout of AI technology and AI hardware, among others. Founded in April 2015 by Xi Zhou, a former professor at Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, CloudWalk Technology produces facial recognition terminals, facial scanning door entry and infrared binoculars scanning machines

Indonesia’s online community platform KASKUS has invested in NLP company Prosa.ai – Oct. 8, 2018 (Yahoo)

  • KASKUS will user the Text Processing Tool by Prosa.ai to filter hoax and negative news in its forum in an effort to bring more positive contents to Kaskuser, its loyal users.
  • The company that was founded just this year focuses on Text & Speech-based Processing Tools with a purpose to imitate human ability in analysing text in a long form document or a transcription. Text can be sourced from comments, surveys, publishing, response, and others to become an insight for business


Hyundai Mobis And Tata Elxsi Join Forces to Develop Driverless Cars – Oct. 12, 2018 (Car and Bike)

  • Hyundai Mobis Technical Centre India has tied-up with Tata Elxsi to develop a simulation tool that will play a key role in development of driverless cars. Hyundai Mobis is a public South Korean car parts company which is the parts and service division of Hyundai Motors and also supply parts to Hyundai owned Genesis Motors and Kia Motors.
  • The South Korean brand has formed partnership with Tata Elxsi to develop a synthetic scene generator tool. The tool that will be developed with the expertise of both the companies is claimed that can deal every real world scenario a vehicle could encounter

IBM and NVIDIA Collaborate to Expand Open Source Machine Learning Tools for Data Scientists – Oct. 10, 2018 (HPCWire)

  • IBM today announced that it plans to incorporate the new RAPIDS open source software into its enterprise-grade data science platform for on-premises, hybrid, and multicloud environments. With IBM’s vast portfolio of deep learning and machine learning solutions, it is best positioned to bring this open-source technology to data scientists regardless of their preferred deployment model
  • RAPIDS will help bring GPU acceleration capabilities to IBM offerings that take advantage of open source machine learning software including Apache Arrow, Pandas and scikit-learn. Immediate, wide ecosystem support for RAPIDS comes from key open-source contributors including Anaconda, BlazingDB, Graphistry, NERSC, PyData, INRIA, and Ursa Labs

Research / studies:

MIT Creates AI that Predicts Depression from Speech – Oct. 14, 2018 (Psychology Today)

  • An innovative Massachusetts Institute of Technology (MIT) research team consisting of Tuka Alhanai and James Glass at CSAIL (Computer Science and Artificial Intelligence Laboratory), and Mohammad Ghassemi at IMES (Institute for Medical Engineering and Science) discovered a way for AI to detect depression in individuals through identifying patterns in natural conversation
  • The MIT researchers developed a neural-network AI model based that could predict depression based on identifying speech patterns from audio and text transcriptions from interviews. Using a data set from 142 recorded patient interviews, the team aimed to model sequences for depression detection. The researchers included experiments in context-free modeling, weighted modeling, and sequence modeling

Uber-funded research project pushes new ways to measure safety of driverless cars – Oct. 11, 2018 (Washington Post)

  • A new study by the Rand Corp., funded by Uber’s autonomous vehicle division and released Thursday, tries to map out what independent tests of driverless safety might look like and how they might be implemented
  • One key element, the authors say, would be trying to define and gauge “roadmanship,” a 21st-century riff on good citizenship and driving behavior by robotic cars

Intel launches AI research center in Israel – Oct. 9, 2018 (VentureBeat)

  • Intel has announced plans to open a joint AI research center in Haifa, Israel. The center will be run in a collaboration between Intel AI Research and Technion – Israel Institute of Technology. Outside of Intel operations on the West Coast in the United States, Intel AI Research has offices in Munich, Germany; Bangalore, India; Beijing, China; and one in Haifa that has often employed Technion graduates

Government / policy:

Tencent executive urges Europe to focus on ethical uses of artificial intelligence – Oct. 14, 2018 (The Telegraph)

  • Chinese tech giant Tencent has urged European companies to focus on ethical applications of artificial intelligence, leaving higher-risk ventures to the US and China
  • Speaking at a conference in Helsinki, Finland, David Wallerstein, Tencent’s chief exploration officer, said he was encouraging the European Union to “embrace AI and deploy it in the areas that would have a maximum benefit for human life, even if that technology isn’t competitive to take on an American or Chinese market”

Self-driving taxis hit the road in Dubai – Oct. 14, 2018 (Mashable)

  • The RTA said the cars go up about 20 mph and fit four passengers. Up front is a safety driver who can take over the car if needed. In promotional materials the cars are depicted without anyone in the front seat for a truly driverless experience, hinting at what the transportation agency could introduce later. The cars are modified Mercedes-Benz sedans with cameras, sensors, and LiDAR system built into the car
  • The taxis come after Dubai tested self-driving passenger drones for an air-based taxi system. That service hasn’t taken off yet for the general public to use

Arizona is creating an autonomous vehicle research institute – Oct. 12, 2018 (Engadget)

  • “The Institute for Automated Mobility will bring together global industry leaders, a public sector team and the brightest minds in academia, focused on advancing all aspects of automated vehicle science, safety and policy,” Governor Doug Ducey said in a statement. “Arizona is committed to providing the leadership and knowledge necessary to integrate these technologies into the world’s transportation systems.”
  • Once completed, IAM will include research facilities, simulation labs and a safety test track that can accomodate different road and traffic configurations. Researchers will explore a number of topics related to autonomous technology, such as safety, accident liability and compensation models. And a Traffic Incident Management center will “integrate law enforcement and first responders with automated vehicle technologies unlike any other location in the country,” according to the Arizona Commerce Authority.

Google will not bid for the Pentagon’s $10B cloud computing contract, citing its “AI Principles” – Oct. 8, 2018 (TechCrunch)

  • Google has dropped out of the running for JEDI, the massive Defense Department cloud computing contract potentially worth $10 billion. In a statement to Bloomberg, Google said that it decided not to participate in the bidding process, which ends this week, because the contract may not align with the company’s principles for how artificial intelligence should be used
  • In statement to Bloomberg, Google spokesperson said “We are not bidding on the JEDI contract because first, we couldn’t be assured that it would align with our AI Principles. And second, we determined that there were portions of the contract that were out of scope with our current government certifications,” adding that Google is still “working to support the U.S. government with our cloud in many ways.”

DOT seeking feedback on new autonomous vehicle guidance – Oct. 8, 2018 (CCJ)

  • Preparing for the Future of Transportation: Automated Vehicles 3.0 (AV 3.0), builds on Automated Driving Systems 2.0: A Vision for Safety and expands the scope by providing new multi-modal guidance. One key update involving the trucking industry in AV 3.0 is that the DOT will broaden the interpretation of “driver” to no longer assume a human is always driving a commercial vehicle