Artificial intelligence is rapidly changing many aspects of how we work and live. (How many stories did you read last week about self-driving cars and job-stealing robots? Perhaps your holiday shopping involved some AI algorithms, as well.) But despite the constant flow of news, many misconceptions about AI remain.
AI doesn’t think in our sense of the word at all, Scriffignano explains. “In many ways, it’s not really intelligence. It’s regressive.”
IT leaders should make deliberate choices about what AI can and can’t do on its own. “You have to pay attention to giving AI autonomy intentionally and not by accident,”
5,173 Comments
Tomi Engdahl says:
Enabling Embedded Vision Neural Network DSPs
https://semiengineering.com/enabling-embedded-vision-neural-network-dsps/
Why DSPs excel at embedded vision neural networks.
Tomi Engdahl says:
Improving Library Characterization with Machine Learning
https://semiengineering.com/improving-library-characterization-with-machine-learning/
New approaches for fast and accurate characterization and validation.
Efficient and accurate library characterization is a critical step in full-chip or block-level design flows because it ensures that all library elements perform to specification under all intended operating conditions. However, traditional library characterization and validation have become increasingly expensive in terms of computation and engineering effort, due to complexity and the amount of characterized data. As characterization needs exceed the scalability of traditional methodologies, the risk of schedule delays, incomplete verification of characterized results, and re-spins due to chip failures increases.
https://www.mentor.com/products/ic_nanometer_design/resources/overview/improving-library-characterization-with-machine-learning-df813b73-150e-4be4-8fbe-21f9e7a05044?cmpid=10168
Tomi Engdahl says:
Machine Learning at DesignCon 2019: Get the Details
https://www.designnews.com/content/machine-learning-designcon-2019-get-details/45682803959962?ADTRK=UBM&elq_mid=6853&elq_cid=876648
New methods of solving SI/PI problems using machine learning and artificial intelligence include Bayesian learning, surrogate modeling, and recurrent neural networks.
Machine learning (ML) and artificial intelligence (AI) have been taking the technology industries by storm for the past few years, from self-driving cars and acrobatic robots to a champion computer chess player. Yet in the area of hardware and systems design, there was little push to utilize these amazing technologies to enhance the design process. With the exception of integrated circuit layout tools, there was very little hardware and system design methodologies with ML and AI as their main ingredient. This was particularly true in the area of high-speed signal integrity (SI) and power integrity (PI) analysis.
Tomi Engdahl says:
https://semiengineering.com/system-bits-dec-18/
Language processing is a leading area in artificial intelligence research, Stanford University reports. “We’re trying to inform the conversation about artificial intelligence with hard data,” says Yoav Shoham, professor of computer science, emeritus, adding, “Language is the ultimate frontier of AI research because you can express any thought or idea in language. It’s as rich as human thinking.” The university just issued its second annual report on AI technology, which is available here. One section of the report is devoted to progress in the field of natural language processing for AI purposes.
http://cdn.aiindex.org/2018/AI%20Index%202018%20Annual%20Report.pdf
Tomi Engdahl says:
A Stanford-led survey of trends in artificial intelligence finds advances in working with human languages, global reach
https://news.stanford.edu/press-releases/2018/12/12/artificial-intelage-global-reach/
The group’s report helps investors and governmental agencies and provides updates for people whose lives will be affected by new developments in AI.
Tomi Engdahl says:
Will AI Drive Scaling Forward?
https://semiengineering.com/will-ai-drive-scaling-forward/
There may be a new application that demands more density and processing power.
The almost ubiquitous rollout of AI and its offshoots—machine learning, deep learning, neural nets of all types—will require significantly more processing power as the amount of data that needs to be processed continues to grow by orders of magnitude. What isn’t clear yet is how that will affect semiconductor manufacturing or how quickly that might happen.
Tomi Engdahl says:
What is machine vision, and how can it help?
https://www.controleng.com/articles/what-is-machine-vision-and-how-can-it-help/
Understanding how machine vision works will help you see if machine vision will clear up specific application difficulties in manufacturing or processing.
Tomi Engdahl says:
Deep Learning Hardware: FPGA vs. GPU
https://semiengineering.com/deep-learning-hardware-fpga-vs-gpu/
While GPUs are well-positioned in machine learning, data type flexibility and power efficiency are making FPGAs increasingly attractive.
FPGAs or GPUs, that is the question.
Tomi Engdahl says:
AI Market Ramps Everywhere
https://semiengineering.com/what-is-artificial-intelligence/
The term creates hope for some, fear for others, and confusion for all.
Tomi Engdahl says:
https://www.uusiteknologia.fi/2018/12/18/nvidia-tuo-tekoalyn-robotiikkaan/
Tomi Engdahl says:
This project is mapping every solar panel in the country using machine learning
https://techcrunch.com/2018/12/19/this-project-is-mapping-every-solar-panel-in-the-country-using-machine-learning/?sr_share=facebook&utm_source=tcfbpage
Tomi Engdahl says:
In the Quest for General Intelligence, AIs Are Chasing Chickens in Minecraft
https://spectrum.ieee.org/robotics/artificial-intelligence/in-the-quest-for-general-intelligence-ais-are-chasing-chickens-in-minecraft
If artificial intelligence (AI) agents are to become real players in society, using their machine abilities to complement our human strengths, they must first become players in the video game of Minecraft. And to prove themselves in Minecraft, they must work together to capture animals in a maze, build towers of blocks, and hunt for treasure while fighting off skeletons.
That, anyway, is the premise of a competition organized by Microsoft, Queen Mary University of London, and crowdAI (a platform for data-science challenges). Next month, the organizers will announce the winner—the team that created an AI that could best observe its Minecraft environment, determine which of three missions it had to accomplish, and then collaborate with another AI agent to carry out that mission.
Tomi Engdahl says:
Facebook has a plan to track you offline
https://www.slashgear.com/facebook-has-an-idea-to-track-you-offline-17558170/
This week a plan for a future Facebook app feature was revealed in a patent for “Office Trajectories” by the USPTO. In this patent, Facebook detailed a method for determining the current location of an individual – even when their phone is turned off and/or their GPS is deactivated. How might Facebook achieve this, you might scream? They’ll just use all the information you’ve already given them, jam it all in a computer with Machine Learning, and spit out the most likely location – it’s easy, really!
we can go right on ahead and file the fact that this exists under our “We’re Being Tracked” file. Yes, Uncle Jimmy, you are being tracked, just like you always thought you were – but not by the government. You’re being tracked by the companies you use to share and communicate with friends and family.
Facebook has filed patents to predict our future locations
https://nakedsecurity.sophos.com/2018/12/14/facebook-has-filed-patents-to-predict-our-future-locations/
Facebook filed a patent, titled “Offline Trajectories,” last week in which it proposes predicting users’ “location trajectories” – in other words, where we’re likely headed. Knowing when we’re about to hurtle into a no-WiFi-connection limbo means Facebook can “prefill” our phones with content and ads.
It knows enough to know a lot more
Tomi Engdahl says:
Forget Cats, This Neural Network Spots Solar Panels
https://spectrum.ieee.org/view-from-the-valley/energy/renewables/forget-cats-this-neural-network-spots-solar-panels
Tomi Engdahl says:
THE RISKS AND REWARDS OF ARTIFICIAL INTELLIGENCE
https://www.sas.com/sas/offers/18/risk-and-rewards-of-artificial-intelligence.html
A Harvard Business Review Analytics Services Insights Report
How will AI, Machine Learning and advanced algorithms impact our lives, our jobs and the economy?
Harvard Business Review Analytic Services published a series of expert articles
Tomi Engdahl says:
Can You Spot Which Of These Faces Are Real?
https://www.iflscience.com/technology/can-you-spot-which-of-these-faces-are-real/
Researchers at AI company Nvidia have modified Generative Adversarial Network (GAN) technology, which can create fake human faces so photorealistic that actual humans are unable to distinguish them from a genuine photograph.
Tomi Engdahl says:
Breaking CAPTCHA Using Machine Learning in 0.05 Seconds
https://medium.com/mlmemoirs/breaking-captcha-using-machine-learning-in-0-05-seconds-9feefb997694
Machine learning model breaks CAPTCHA systems on 33 highly visited websites, concept is based on GANs.
Tomi Engdahl says:
Fujitsu Plans to Support Professional Judges With Lidar and AI at Gymnastics Meets
https://spectrum.ieee.org/tech-talk/computing/software/fujitsu-plans-to-support-professional-judges-with-lidar-and-ai-at-gymnastics-meets
To ensure a fair competition, organizers typically rely on highly trained human eyes. Now, an international gymnastics association has agreed to incorporate lidar-based technology developed by Fujitsu into future competitions to help judges assess performances. The organization plans to fully automate scoring by 2020.
Tomi Engdahl says:
Zachary Fryer-Biggs / Wired:
Sources describe DOD’s plan to win “hearts and minds” in Silicon Valley, amid frustration and anger inside the Pentagon after Google pulled out of Project Maven — THE AMERICAN MILITARY is desperately trying to get a leg up in the field of artificial intelligence …
Inside the Pentagon’s Plan to Win Over Silicon Valley’s AI Experts
https://www.wired.com/story/inside-the-pentagons-plan-to-win-over-silicon-valleys-ai-experts/
The American military is desperately trying to get a leg up in the field of artificial intelligence, which top officials are convinced will deliver victory in future warfare. But internal Pentagon documents and interviews with senior officials make clear that the Defense Department is reeling from being spurned by a tech giant and struggling to develop a plan that might work in a new sort of battle—for hearts and minds in Silicon Valley.
The battle began with an unexpected loss. In June, Google announced it was pulling out of a Pentagon program—the much-discussed Project Maven—that used the tech giant’s artificial intelligence software. Thousands of the company’s employees had signed a petition two months earlier calling for an end to its work on the project, an effort to create algorithms that could help intelligence analysts pick out military targets from video footage.
The reason for the Pentagon’s anxiety is clear: It wants a smooth path to use artificial intelligence in weaponry of the future, a desire already backed by the promise of several billion dollars to try to ensure such systems are trusted and accepted by military commanders, plus billions more in expenditures on the technologies themselves.
Tomi Engdahl says:
Bärí a. Williams / Fast Company:
The First Step Act, which seeks to reform the federal criminal justice system, risks perpetuating racial and class disparities due to its reliance on algorithms
The prison-reforming First Step Act has a critical software bug
https://www.fastcompany.com/90284823/the-first-step-acts-reliance-on-algorithms-is-a-misstep
The bipartisan bill has been hailed as a triumph, but its reliance on algorithms might only reinforce existing disparities.
The road to hell is paved with good intentions. Often in a rush of excitement and eagerness to solve a problem, or to alleviate a particularly thorny issue, people will champion a solution without being aware of all of the possibilities and potential pitfalls, intentional or otherwise, or blatantly ignore the risks because the reward seems so great.
The First Step Act has lofty and noble goals. This piece of legislation seeks to reform the criminal justice system at the federal leve
What is most notable is the use of technology in the bill. Though the use of technology in policing is not uncommon , predictive policing is used to discern where to dispatch officers at any given time, and facial recognition is spreading across public space, the use of artificial intelligence in determining the fate of those already imprisoned is new. But some recent events should give citizens pause about the enthusiasm of legislators to apply technology in prison reform.
Red flags abound. During a congressional hearing this week, Google CEO Sundar Pichai detailed the perceived privacy issues and biases of their products as noted by lawmakers.
What is most notable is the use of technology in the bill. Though the use of technology in policing is not uncommon , predictive policing is used to discern where to dispatch officers at any given time, and facial recognition is spreading across public space, the use of artificial intelligence in determining the fate of those already imprisoned is new. But some recent events should give citizens pause about the enthusiasm of legislators to apply technology in prison reform.
Red flags abound. During a congressional hearing this week, Google CEO Sundar Pichai detailed the perceived privacy issues and biases of their products as noted by lawmakers.
While technically true—algorithms do not have political biases—the people who create them do, as well as gender, religious, and racial biases. Algorithms are only as good as the data they are fed, which is only as diverse and inclusive as those writing the code.
Furthermore, the AI systems used in criminal justice lack transparency: Who is fact checking the fact checkers? Who is setting the parameters of what is deemed relevant information to include in the decision-making process? Who is checking the rate of diverse representation in datasets to ensure they aren’t skewed, and if there is parity or equity in the information shared and diverse perspectives of data and research being used as the basis for any sort of dataset? The lack of transparency offered regarding decisions made through AI systems leads to a lack of accountability, as there is no way to thoroughly audit the information and the process. Essentially, without being able to properly audit algorithms used in sentencing, we aren’t aware of the possibly skewed outcomes, nor can we correct it sufficiently.
Though well intentioned, the bill is a wolf in sheep’s clothing, particularly when you add the technology component. Laws shouldn’t be written out of fear, but in a place of strength.In this post Cambridge Analytica world, there is a lot of fearmongering. Technology isn’t always the answer, particularly when ethics are paramount
Tomi Engdahl says:
Matt Sheehan / MIT Technology Review:
A history of Google’s relationship with China from its first foray in 2006 to Project Dragonfly, and how Google hopes to reenter China in the era of AI
How Google took on China—and lost
https://www.technologyreview.com/s/612601/how-google-took-on-china-and-lost/
Google’s first foray into Chinese markets was a short-lived experiment.
Observers talk as if the decision about whether to reenter the world’s largest market is up to Google: will it compromise its principles and censor search the way China wants? This misses the point—this time the Chinese government will make the decisions.
Tomi Engdahl says:
10 Ways Machine Learning Is Revolutionizing Sales
https://www.forbes.com/sites/louiscolumbus/2018/12/26/10-ways-machine-learning-is-revolutionizing-sales/#36b712663fd1
Sales teams adopting AI are seeing an increase in leads and appointments of more than 50%, cost reductions of 40%–60%, and call time reductions of 60%–70% according to the Harvard Business Review article Why Salespeople Need to Develop Machine Intelligence.
By 2020, 30% of all B2B companies will employ AI to augment at least one of their primary sales processes according to Gartner.
High-performing sales teams are 4.1X more likely to use AI and machine learning applications than their peers according to the State of Sales published by Salesforce.
Tomi Engdahl says:
Tekoäly on kupla, joka puhkesi viimeksi 1990-luvulla
https://www.tivi.fi/Kaikki_uutiset/tekoaly-on-kupla-joka-puhkesi-viimeksi-1990-luvulla-6753698
Curious AI:n perustajiin kuuluvan Timo Haanpään mukaan hype tekoälyn ympärillä laantuu. Nykyisellään tekoäly ja ihminen muodostavat yhdessä huonon kyborgin.
Miksi tekoälystä puhutaan juuri nyt niin paljon?
”Alamme lopulta saada asioita laskettavaan muotoon. Laskentakapasiteetin moorelainen kasvu tuo suurteholaskennan taskuun. Tekoälykön ei myöskään enää tarvitse olla velho: parhaimmillaan toteutuksen voi poimia GitHubista [koodinjakopalvelu] tajuamatta mistään mitään.”
Onko tekoäly kupla ja milloin se puhkeaa?
”Se on ihan täysin kupla, joka puhkesi viimeksi isosti 1990-luvulla. 2000-luvulla ilmaa pääsi vielä vähän pallosta. 2010-luvulle tultaessa syväoppiminen räjäytti pankin. Viimeisin hype saavutti huippunsa vuonna 2016, minkä jälkeen on taas lasketeltu kohti tekoälytalvea.”
Tomi Engdahl says:
Hardware-Software Co-Design Approach Could Make Neural Networks Less Power Hungry
https://ucsdnews.ucsd.edu/pressrelease/hardware_software_co_design_approach_could_make_neural_networks_less_power_hungry
A team led by the University of California San Diego has developed a neuroinspired hardware-software co-design approach that could make neural network training more energy-efficient and faster.
Kuzum and her lab teamed up with Adesto Technologies to develop hardware and algorithms that allow these computations to be performed directly in the memory unit, eliminating the need to repeatedly shuffle data.
Tomi Engdahl says:
2018: The Year Machine Intelligence Arrived in Cybersecurity
https://www.darkreading.com/network-and-perimeter-security/2018-the-year-machine-intelligence-arrived-in-cybersecurity/d/d-id/1333556
Machine intelligence, in its many forms, began having a significant impact on cybersecurity this year – setting the stage for growing intelligence in security automation for 2019.
Tomi Engdahl says:
The Welfare State Is Committing Suicide by Artificial Intelligence
https://foreignpolicy.com/2018/12/25/the-welfare-state-is-committing-suicide-by-artificial-intelligence/
Denmark is using algorithms to deliver benefits to citizens—and undermining its own democracy in the process.
As a philosophy of government, liberalism is premised on the belief that the coercive powers of public authorities should be used in service of individual freedom and flourishing, and that they should therefore be constrained by laws controlling their scope, limits, and discretion. That is the basis for historic liberal achievements such as human rights and the rule of law, which are built into the infrastructure of the Scandinavian welfare state.
Yet the idea of legal constraint is increasingly difficult to reconcile with the revolution promised by artificial intelligence and machine learning—specifically, those technologies’ promises of vast social benefits in exchange for unconstrained access to data and lack of adequate regulation on what can be done with it.
Tomi Engdahl says:
A majority of doctors in a recent survey did not think that AI will supplant them when it comes to diagnoses (68%), referrals (61%), treatment plans (61%), or empathy (94%).
AI Won’t Replace Us, Docs Say
https://spectrum.ieee.org/the-human-os/biomedical/ethics/ai-wont-replace-us-docs-say
Each time AI is pitted against doctors in a medical task—and we’ve covered a lot of them—one question inevitably bubbles to the surface: Will AI replace physicians?
If you talk to AI experts or Silicon Valley investors, the answer is often yes. But, up to this point, no one really asked the doctors.
In a first-of-its-kind national survey of primary care physicians in the UK, the overwhelming majority of doctors said, “No,” regardless of their age or gender. The docs were skeptical that artificial intelligence will replace them in any of six key medical tasks—except paperwork. They’re happy to let bots do the filing.
Tomi Engdahl says:
The Most Terrifying Thought Experiment of All Time
https://slate.com/technology/2014/07/rokos-basilisk-the-most-terrifying-thought-experiment-of-all-time.html
Why are techno-futurists so freaked out by Roko’s Basilisk?
One day, LessWrong user Roko postulated a thought experiment: What if, in the future, a somewhat malevolent AI were to come about and punish those who did not do its bidding?
The LessWrong community is concerned with the future of humanity, and in particular with the singularity—the hypothesized future point at which computing power becomes so great that superhuman artificial intelligence becomes possible, as does the capability to simulate human minds, upload minds to computers, and more or less allow a computer to simulate life itself. The term was coined in 1958 in a conversation between mathematical geniuses Stanislaw Ulam and John von Neumann, where von Neumann said, “The ever accelerating progress of technology … gives the appearance of approaching some essential singularity in the history of the race beyond which human affairs, as we know them, could not continue.” Futurists like science-fiction writer Vernor Vinge and engineer/author Kurzweil popularized the term, and as with many interested in the singularity, they believe that exponential increases in computing power will cause the singularity to happen very soon—within the next 50 years or so.
Tomi Engdahl says:
AI used for UI testing:
Automated visual
UI testing
https://applitools.com/landing/free-account-qa/?utm_source=paid-social&utm_medium=facebook&utm_campaign=remarketing-video&utm_content=free-account&utm_term=clever-zebo
For a visually perfect digital experience.
Applitools helps Test Automation engineers, DevOps, and FrontEnd Developers continuously test and validate visually perfect mobile, web, and native apps.
Tomi Engdahl says:
Andrew Tarantola / Engadget:
AI image processing for computer vision, facial recognition, image generation, and other applications saw marked improvements in 2018
2018 is the year AI got its eyes
Art and commerce may never be the same.
https://www.engadget.com/2018/12/29/2018-is-the-year-ai-got-its-eyes/
Tomi Engdahl says:
7 Predictions for AI in 2019
https://www.eetimes.com/author.asp?section_id=36&doc_id=1334117
Artificial intelligence is a vast field with many unknowns, but it’s not hard to predict a few things that will or should happen in 2019 with the part of it that is deep learning.
We’ve gotten sloppy in our language. It’s convenient to use AI as shorthand for deep learning — and it gets good hits in headlines. But these days, general AI — machines learning on their own like curious humans browsing in a bookstore — is still more science fiction than science.
What’s spreading like wildfire through the internet these days are deep neural networks, a special case of AI based on processes typically initiated by people. The ability of deep-learning techniques to recognize patterns in images, speech, and other areas — often faster than people can — has opened a door to a whole new direction in computing. Where this goes long-term is anyone’s guess.
Tomi Engdahl says:
Researchers Explore Emerging Memories for AI
https://www.eetimes.com/document.asp?doc_id=1334122
Resistive random access memory (ReRAM) and other emerging memory technologies have been getting a lot of attention in the past year as semiconductor companies look for ways to more efficiently deal with the requirements of artificial intelligence and neuromorphic computing.
At the International Electron Devices Meeting (IEDM) in San Francisco earlier this month, there were several papers presented that dealt with using emerging memory in neomorphic computing from companies the likes of IBM and various universities.
“The neuromorph crowd is excited about this type of thing,” said Jim Handy, a veteran memory market watcher who is principal analyst at Objective Analysis. “I wouldn’t say that any [the emerging memory technologies} stands out. They all have something. The question is who is going to get something meaningful to the market first.”
Tomi Engdahl says:
Drew Harwell / Washington Post:
How deepfakes, AI-generated videos that graft a person’s face onto another’s body, have been weaponized to harass and humiliate their subjects, mostly women — “Deepfake” creators are making disturbingly realistic, computer-generated videos with photos taken from the Web, and ordinary women are suffering the damage.
Fake-porn videos are being weaponized to harass and humiliate women: ‘Everybody is a potential target’
https://www.washingtonpost.com/technology/2018/12/30/fake-porn-videos-are-being-weaponized-harass-humiliate-women-everybody-is-potential-target/?utm_term=.7e0e0fa378f1
‘Deepfake’ creators are making disturbingly realistic, computer-generated videos with photos taken from the Web, and ordinary women are suffering the damage
The video showed the woman in a pink off-the-shoulder top, sitting on a bed, smiling a convincing smile.
It was her face. But it had been seamlessly grafted, without her knowledge or consent, onto someone else’s body: a young pornography actress, just beginning to disrobe for the start of a graphic sex scene. A crowd of unknown users had been passing it around online.
She felt nauseated and mortified: What if her co-workers saw it? Her family, her friends? Would it change how they thought of her? Would they believe it was a fake?
“I feel violated — this icky kind of violation,”
Tomi Engdahl says:
Andrew Tarantola / Engadget:
AI image processing for computer vision, facial recognition, image generation, and other applications saw marked improvements in 2018
2018 is the year AI got its eyes
Art and commerce may never be the same.
https://www.engadget.com/2018/12/29/2018-is-the-year-ai-got-its-eyes/
Computer scientists have spent more than two decades teaching, training and developing machines to see the world around them. Only recently have the artificial eyes begun to match (and occasionally exceed) their biological predecessors. 2018 has seen marked improvement in two areas of AI image processing: facial-recognition technology in both commerce and security, and image generation in — of all fields — art.
In September of this year, a team of researchers from Google’s DeepMind division published a paper outlining the operation of their newest Generative Adversarial Network. Dubbed BigGAN, this image-generation engine leverages Google’s massive cloud computing power to create extremely realistic images. But, even better, the system can be leveraged to generate dreamlike, almost nightmarish, visual mashups of objects, symbols and virtually anything else you train the system with. Google has already released the source code into the wilds of the internet and is allowing creators from anywhere in the world to borrow its processing capabilities to use the system as they wish.
“I’ve been really excited by all of the interactive web demos that people have started to turn these algorithms into,”
Tomi Engdahl says:
Picovoice Puts Smarts Offline in 512K of Memory
https://hackaday.com/2019/01/01/picovoice-puts-smarts-offline-in-512k-of-memory/
We live in the future. You can ask your personal assistant to turn on the lights, plan your commute, or set your thermostat. If they ever give Alexa sudo, she might be able to make a sandwich. However, you almost always see these devices sending data to some remote server in the sky to do the analysis and processing. There are some advantages to that, but it isn’t great for privacy as several recent news stories have pointed out. It also doesn’t work well when the network or those remote servers crash — another recent news story. But what’s the alternative? If Picovoice has its way, you’ll just do all the speech recognition offline.
There’s an ARM board
The libraries are apparently quite portable and the Linux and Raspberry Pi versions are already open source. The company says they will make other platforms available in upcoming releases and claim to support ARM Cortex-M, Cortex-A, Android, Mac, Windows, and WebAssembly.
We imagine that’s true because you can see the ARM version working in the video and there are browser-based demos on their website. They say the code is in ANSI C and uses fixed point math to do all the neural network magic, so the code should be portable.
The libraries on GitHub include:
Rhino – Speech to intent (in other words, do something in response to a spoken command)
Porcupine – Wake word detection
Cheetah – Speech to text
https://picovoice.ai/
Tomi Engdahl says:
This million-core supercomputer inspired by the human brain breaks all the rules
https://www.zdnet.com/article/this-million-core-supercomputer-inspired-by-the-human-brain-breaks-all-the-rules/?ftag=TRE-03-10aaa6b&bhid=21467808899609979249758010609979
SpiNNaker’s spiking neural network mimics the human brain, and could fuel breakthroughs in robotics and health.
Tomi Engdahl says:
THE NEURAL NETWORK ZOO
http://www.asimovinstitute.org/neural-network-zoo/
With new neural network architectures popping up every now and then, it’s hard to keep track of them all. Knowing all the abbreviations being thrown around (DCIGN, BiLSTM, DCGAN, anyone?) can be a bit overwhelming at first.
So I decided to compose a cheat sheet containing many of those architectures. Most of these are neural networks, some are completely different beasts. Though all of these architectures are presented as novel and unique, when I drew the node structures… their underlying relations started to make more sense.
Tomi Engdahl says:
This clever AI hid data from its creators to cheat at its appointed task
https://techcrunch.com/2018/12/31/this-clever-ai-hid-data-from-its-creators-to-cheat-at-its-appointed-task/?sr_share=facebook&utm_source=tcfbpage
Tomi Engdahl says:
Finland’s grand AI experiment
https://www.politico.eu/article/finland-one-percent-ai-artificial-intelligence-courses-learning-training/?utm_medium=paid&utm_campaign=Julkaisu%3A+“”The+idea+has+a+simple%2C+Nordic+ring+to+it%3A+Start…”&utm_source=facebook&hsa_ad=6113236820674&hsa_ver=3&hsa_acc=273790800&hsa_src=fb&hsa_net=facebook&hsa_grp=6113236820074&hsa_cam=6113236819074
Inside Finland’s plan to train its population in artificial intelligence.
Tomi Engdahl says:
Tekoäly voittaisi jo nyt ihmisen oikeassa sodassa – ja sen ymmärtäminen on olennaista, kun Suomi tekee päätöksen historiallisesta hävittäjähankinnasta
https://yle.fi/uutiset/3-10552743
Tomi Engdahl says:
Alexis C. Madrigal / The Atlantic:
A machine-learning model in Flint that once helped predict lead pipe locations with 70% accuracy was abandoned due to political decisions to excavate everywhere
How a Feel-Good AI Story Went Wrong in Flint
https://www.theatlantic.com/technology/archive/2019/01/how-machine-learning-found-flints-lead-pipes/578692/
A machine-learning model showed promising results, but city officials and their engineering contractor abandoned it.
More than a thousand days after the water problems in Flint, Michigan, became national news, thousands of homes in the city still have lead pipes, from which the toxic metal can leach into the water supply. To remedy the problem, the lead pipes need to be replaced with safer, copper ones. That sounds straightforward, but it is a challenge to figure out which homes have lead pipes in the first place. The City’s records are incomplete and inaccurate. And digging up all the pipes would be costly and time-consuming.
That’s just the kind of problem that automation is supposed to help solve.
The artificial intelligence was supposed to help the City dig only where pipes were likely to need replacement. Through 2017, the plan was working. Workers inspected 8,833 homes, and of those, 6,228 homes had their pipes replaced—a 70 percent rate of accuracy.
AECOM discarded the machine-learning model’s predictions, which had guided excavations. And facing political pressure from some residents, Weaver demanded that the firm dig across the city’s wards and in every house on selected blocks, rather than picking out the homes likely to have lead because of age, property type, or other characteristics that could be correlated with the pipes.
The declining success of the pipe-replacement program has caused critics of the City to raise the alarm.
“It’s the number of lead pipes removed that matters, not the number of holes dug,”
Before things got ugly, the effort to pull the lead pipes out of the ground was shaping up to be a high-tech feel-good story.
Many cities share the lead-pipe problem and the informational obstacles layered atop it. The decay of infrastructure built decades ago is not only in the metal, but in the data cataloging that lets the city’s government and residents understand the state of the water system. For all the talk of “smart” cities, the real state of play in many older places is that no one even thinks of these things until there’s a disaster. People have been saying “America is 1,000 Flints” since the city was booming, and it is still true. Just as there are thousands of lead service lines in Flint, there are something like 6 million lead service lines in America.
Some basic things were known about the lead-pipe distribution: The pipes were most likely to be found in postwar homes, built when Flint experienced major expansions, and least likely to be found in newer homes.
Tomi Engdahl says:
http://www.etn.fi/index.php/13-news/8900-tutkijat-kehittivat-oppivan-konenaon
Tomi Engdahl says:
DARPA wants to build an AI to find the patterns hidden in global chaos
https://techcrunch.com/2019/01/07/darpa-wants-to-build-an-ai-to-find-the-patterns-hidden-in-global-chaos/?sr_share=facebook&utm_source=tcfbpage
That most famous characterization of the complexity causality, a butterfly beating its wings and causing a hurricane on the other side of the world, is thought-provoking but ultimately not helpful. What we really need is to look at a hurricane and figure out which butterfly caused it — or perhaps stop it before it takes flight in the first place. DARPA thinks AI should be able to do just that.
A new program at the research agency is aimed at creating a machine learning system that can sift through the innumerable events and pieces of media generated every day and identify any threads of connection or narrative in them. It’s called KAIROS: Knowledge-directed Artificial Intelligence Reasoning Over Schemas.
Tomi Engdahl says:
Americans want to regulate AI but don’t trust anyone to do it
https://www.technologyreview.com/s/612734/americans-want-to-regulate-ai-but-dont-trust-anyone-to-do-it/
The public thinks that AI is likely to cause more harm than good, a new report has shown.
In 2018, several high-profile controversies involving AI served as a wakeup call for technologists, policymakers, and the public. The technology may have brought us welcome advancements across many fields, but it can also fail catastrophically when built shoddily or applied carelessly.
It’s hardly a surprise, then, that Americans have mixed support for the continued development of AI and overwhelmingly agree that it should be regulated
These are important lessons for policymakers and technologists to consider in the discussion on how best to advance and regulate AI, said Allan Dafoe, director of the Center and co-author of the report. “There isn’t currently a consensus in favor of developing advanced AI, or that it’s going to be good for humanity,” he said. “That kind of perception could lead to the development of AI being perceived as illegitimate or cause political backlashes against the development of AI.”
Tomi Engdahl says:
Marcela Kunova / Journalism.co.uk:
Quartz’s AI Studio has developed bots to help journalists with data analysis tasks, using machine learning to identify patterns in records and more
Quartz AI Studio launches an open-source platform to help journalists use machine learning
https://www.journalism.co.uk/news/quartz-ai-studio-launches-an-open-source-platform-to-help-journalists-use-machine-learning/s2/a732936/
The US publisher offers reporters with no coding skills a set of free tools to help them write better data-driven stories
Tomi Engdahl says:
Lulu Yilun Chen / Bloomberg:
Sources: Alibaba-backed AI startup Megvii, owner of facial recognition developer Face++, is considering an IPO in Hong Kong that could raise as much as $1B
https://www.bloomberg.com/news/articles/2019-01-11/alibaba-backed-ai-startup-megvii-is-said-to-weigh-ipo-this-year
Tomi Engdahl says:
70 percent of clinical geneticists worldwide are using the Face2Gene system, which helps them hone in on possible genetic disorders a patient may have.
Face-Scanning AI Identifies Rare Genetic Disorders
https://spectrum.ieee.org/the-human-os/biomedical/diagnostics/face-scanning-ai-identifies-rare-genetic-disorders
Tomi Engdahl says:
FORMER GOOGLE EXEC: AI WILL REPLACE 40 PERCENT OF JOBS IN 15 YEARS
https://futurism.com/the-byte/google-ai-jobs
Artificial intelligence, whether it’s an application of machine learning or some new technology altogether, is poised to shatter the global economy.
Kai-Fu Lee, a venture capitalist who used to develop artificial intelligence for both Microsoft and Google, told CBS’ 60 Minutes that AI will displace 40 percent of the world’s workers within 15 years.
“I believe [AI] is going to change the world more than anything in the history of mankind,” Lee told CBS. “More than electricity.”
Tomi Engdahl says:
Machine learning leads mathematicians to unsolvable problem
https://www.nature.com/articles/d41586-019-00083-3?utm_source=fbk_nnc&utm_medium=social&utm_campaign=naturenews&sf205797813=1
Simple artificial-intelligence problem puts researchers up against a logical paradox discovered by famed mathematician Kurt Gödel.
A team of researchers has stumbled on a question that is mathematically unanswerable because it is linked to logical paradoxes discovered by Austrian mathematician Kurt Gödel in the 1930s that can’t be solved using standard mathematics.
The mathematicians, who were working on a machine-learning problem, show that the question of ‘learnability’ — whether an algorithm can extract a pattern from limited data — is linked to a paradox known as the continuum hypothesis. Gödel showed that the statement cannot be proved either true or false using standard mathematical language.
Tomi Engdahl says:
Kyle Wiggers / VentureBeat:
Intel’s Nervana, a neural network chip for inference-based workloads, will lack a standard cache hierarchy, and software will directly manage on-chip memory
Intel details Nervana, a neural network chip for inference-based workloads (Updated)
https://venturebeat.com/2019/01/07/intel-announces-nervana-a-neural-network-chip-for-inference-based-workloads/
At a press event at the 2019 Consumer Electronics Show, Intel announced the Nervana Neural Network Processor (NNP-I), an AI chip for inference-based workloads that fits into a GPU-like form factor. It wasn’t an unexpected reveal — Intel announced it was working on a new generation of inference chip way back in 2017 — but its appearance at the press conference today made clear the company’s ambition to capture a large slice of the budding AI chip market.
NNP-I is built on a 10-nanometer Intel process, and will include Ice Lake cores to handle general operations as well as neural network acceleration, Naveen Rao, corporate vice president and general manager of AI at Intel, said in a tweet.