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,”
7,003 Comments
Tomi Engdahl says:
AI Chips Put to Data Center Tests
Microsoft lays out the silicon landscape
https://www.eetimes.com/document.asp?doc_id=1333744
The rubber is about to meet the road in what’s projected to be a $25 billion market for deep-learning accelerators. Data centers are testing multiple chips in the labs now and expect to deploy some next year, probably picking multiple accelerators for different workloads.
So far, Graphcore, Habana, ThinCI, and Wave Computing are in the small subset of 50 vendors who have chips that customers are testing in their labs. Representatives from both groups staked out their positions at the AI Hardware Summit here.
Tomi Engdahl says:
http://www.etn.fi/index.php/13-news/8466-puheohjaus-tulee-crm-jarjestelmiin
Introducing New Einstein Voice – You Talk, AI Listens
https://www.salesforce.com/blog/2018/09/introducing-einstein-voice.html
Salesforce has helped our customers succeed as the #1 CRM company year after year. But the fact is getting granular, real-time data into Salesforce that you can take action on could be easier. What if you could talk to Salesforce, and let artificial intelligence make the best use of that information? Now you can, thanks to Einstein Voice.
Tomi Engdahl says:
http://www.etn.fi/index.php/13-news/8460-puhelimen-tekoaly-moninkertaistuu-ensi-vuonna
Tomi Engdahl says:
How Breakthrough Listen Trained AI to Spot Elusive, Mysterious Radio Bursts
https://spectrum.ieee.org/tech-talk/aerospace/astrophysics/breakthrough-listen-trained-an-ai-to-spot-illusive-mysterious-radio-bursts
Tomi Engdahl says:
7 amazing ways artificial intelligence is used in healthcare
https://www.weforum.org/agenda/2018/09/7-amazing-ways-artificial-intelligence-is-used-in-healthcare
One of the biggest impacts of new technology – and perhaps the most life-changing – will be felt in healthcare.
Diagnosis of illness will be fast and efficient, and medicine will be highly personalised. Wearable technology will be the norm, and we’ll know we are sick before we even get a single symptom. Meanwhile, new drugs will come to market at breakneck speed as clinical trials get faster and more accurate.
Ultimately, we will become our own doctors.
AI is already being used in healthcare, and these seven examples offer a glimpse into our medical future.
Tomi Engdahl says:
Tulevaisuuden yhteiskunta: Energiayhtiö hyödyntää tekoälyä
https://pulse.microsoft.com/fi-fi/business-leadership-fi-fi/manufacturing-fi-fi/fa2-future-society-the-utility-company-turning-up-the-heat-with-ai/?utm_campaign=workproductivity&utm_medium=content_social_ad53_fifi&utm_content=none_FY19_Q1_BDM&utm_source=facebook
Tomi Engdahl says:
Linda Liukkaan kolumni: Kun kasvojentunnistus ei toimi, onko syy ihmisen vai tekoälyn?
https://yle.fi/uutiset/3-10398490
Tomi Engdahl says:
Facebook’s robot coders step into the future of programming
https://nakedsecurity.sophos.com/2018/09/17/facebooks-robot-coders-step-into-the-future-of-programming/
Facebook’s Android app recently became one of the first in the world to run software debugged by Artificial Intelligence (AI).
Called SapFix, the company describes it as an “AI hybrid tool” that can be used in conjunction with the Sapienz automated Android testing tool originally developed by university researchers but taken in-house by Facebook some time ago.
Tomi Engdahl says:
The “real AI crisis,” Kai-Fu Lee says, will come from automation that wipes out whole job sectors, reshaping economies and societies in both the United States and China.
Former Head of Google China Foresees an AI Crisis—and Proposes a Solution
https://spectrum.ieee.org/tech-talk/robotics/artificial-intelligence/former-head-of-google-china-foresees-an-ai-crisis
When the former president of Google China talks about artificial intelligence and its potential to cause global upheaval, people listen. His hope is that enough people will listen to avert catastrophic disruption on three different scales: to the global balance of power, to national economies, and to human beings’ delicate souls.
Tomi Engdahl says:
AI Chips Put to Data Center Tests
Microsoft lays out the silicon landscape
https://www.eetimes.com/document.asp?doc_id=1333744
The rubber is about to meet the road in what’s projected to be a $25 billion market for deep-learning accelerators. Data centers are testing multiple chips in the labs now and expect to deploy some next year, probably picking multiple accelerators for different workloads.
So far, Graphcore, Habana, ThinCI, and Wave Computing are in the small subset of 50 vendors who have chips that customers are testing in their labs. Representatives from both groups staked out their positions at the AI Hardware Summit here.
Tomi Engdahl says:
Cadence AI Core Hits 12 TMACs
Tensilica DNA 100 beats Arm’s ML core
https://www.eetimes.com/document.asp?doc_id=1333740
Cadence announced an inference core with up to four times the multiply-accumulate units and up to 12 times the performance of its Vision C5 launched last year. The DNA 100 core supports sparsity in weights and activations and can prune neural networks to deliver higher levels of performance.
To date, high-end smartphones have led the way in adopting deep learning for inference jobs with handset SoC vendors, such as Mediatek using Cadence’s Vision P6 core. Designers are now working on AI acceleration in SoCs for surveillance cameras, smart speakers, cars, and AR/VR and IoT devices, said Lazaar Louis, a senior director of product management in Cadence’s Tensilica group.
Cadence clocked a 16-nm DNA 100 with 4,000 MACs at up to 2,550 frames/second and up to 3.4 TMACs/W on ResNet-50. A single 16-nm core running at 1 GHz can deliver up to 8 TMACs (12 TMACs using network pruning), and multiple cores can be embedded in an SoC to hit hundreds of TMACs.
Tomi Engdahl says:
Teaching Machines ‘Fairness’
https://www.eetimes.com/document.asp?doc_id=1333748
eaching anyone about “fairness” is a laudable goal.
As humans, we may not necessarily agree on what’s fair. It sometimes depends on the context. Teaching kids to be fair — both at home and in school — is fundamental, but it’s easier said than done. With this in mind, how can we, as a society, communicate the nuances of “being fair” to artificial intelligence (AI) systems?
A team of researchers at IBM Research is taking the first crack at this conundrum. IBM is rolling out a tool kit for developers called “AI Fairness 360.” As part of this effort, IBM is offering businesses a new “cloud-based, bias-detection, and mitigation service” that corporations can use to test and verify how AI-driven systems are behaving.
Tomi Engdahl says:
Deezer’s AI mood detection could lead to smarter song playlists
It pays attention to both the instruments and the lyrics.
https://www.engadget.com/2018/09/23/deezer-ai-song-mood-detection/
Astute listeners know that you can’t gauge a song’s mood solely through the instrumentation or the lyrics, but that’s often what AI has been asked to do — and that’s not much help if you’re trying to sift through millions of songs to find something melancholic or upbeat Thankfully, Deezer’s researchers have found a way to make that AI consider the totality of a song before passing judgment. They’ve developed a deep learning system that gauges the emotion and intensity of a track by relying on a wide variety of data, not just a single factor like the lyrics.
Deezer trained the AI using raw audio signals, linguistic context reconstruction models and a Million Song Dataset that aggregates Last.fm tags describing tunes (such as “calm” or “sad”). The researchers mapped the MSD to Deezer’s library using song metadata, extracting individual words from the lyrics in the process. The result was an 18,644-song database the team could use to both train AI on song moods and to test its theories.
Tomi Engdahl says:
Machine Learning Shifts More Work to FPGAs, SoCs
https://semiengineering.com/machine-learning-shifts-more-work-to-fpgas-socs/
SoC bandwidth, integration expand as data centers use more FPGAs for machine learning.
Tomi Engdahl says:
Renesas to Pitch Baby-step AI for Factories
AI goes inside Renesas MCUs, Dynamically Reconfigurable Processors
https://www.eetimes.com/document.asp?doc_id=1333788
Every company that has pledged its faith to “smart manufacturing” is pledging its hopes for AI.
This brave new world requires a big investment in high-cost AI systems, along with the cost of setting up a “learning” platform and contacting cloud service providers. The grand plan starts with big data collection so that the machine can learn and figure out something previously unknown.
That’s the theory.
In the real world, however, many companies are finding AI hard to implement. Some blame their inexperience in AI, or a shortage of in-house data scientists cable of making the most of AI. Others complain that they have not been able to establish the proof of concept of their installed AI systems. In any case, manufacturers are beginning to realize that AI is not an “if you build it, they will come” deal.
Enter Renesas Electronics.
The Japanese chip company claims a leading position in the global factory automation market. It is proposing “real-time continuous AI” for the world of operational technology (OT). This approach contrasts sharply with “statistical AI,” often pitched by big data companies to promote automation in the world of information technology (IT).
Tomi Engdahl says:
Industrial Robotics Are Expanding Across Multiple Sectors
https://www.designnews.com/automation-motion-control/industrial-robotics-are-expanding-across-multiple-sectors/54664136559474?ADTRK=UBM&elq_mid=5774&elq_cid=876648
Advances in software development techniques and networking technologies have made the installing and assembling of robots faster and less costly.
Robotics has progressed from a world of building blocks and moved to real-world computing. Robots are getting progressively integrated into several types of production lines, which is driving industrial robotics market growth. According to Grand View Research, the industrial robotics market is expected to reach $41.20 billion by 2025. The rising awareness of the benefits of industrial robots—which include cost effectiveness, quality assurance, optimized production efficiency, and safe working conditions in hazardous environments—offers positive opportunities for market growth.
AI + ML + Industrial Robotics
Artificial intelligence (AI) and machine learning (ML) capabilities have gradually made their way into industrial robotics technology, leading to the adoption of collaborative robots or co-bots in various application areas. Co-bots enable direct interaction with a human within a defined collaborative workspace.
Tomi Engdahl says:
Microsoft Office gets smarter
https://techcrunch.com/2018/09/24/microsoft-office-gets-smarter/
Microsoft used its Ignite conference in Orlando, Florida, today to announce a number of new features that are coming to Office 365. Given the company’s current focus on AI, it’s no surprise that most of these new features are powered by AI in one form or another. That means all of your Office apps, on- and offline, will soon become a little bit easier to use and offer you more assistance.
Tomi Engdahl says:
This Health Care AI Loves Terrible Software
https://spectrum.ieee.org/the-human-os/computing/software/this-healthcare-ai-loves-crappy-software
In 2018, we’ve tracked AI edging its way into many corners of health care, primarily in diagnostics but also in patient monitoring, selecting dosing regimens, and drug development.
Well, here’s a new one for you—an AI program that loves crappy health care software.
Lane and a team taught an AI system to use software that already exists in health care just like a human would use it. They named it Olive.
“Olive loves all that crappy software that health care already has,” said Lane. “Olive can look at any software program, any application for the first time she’s ever seen it, and understand how to use it.”
For example, Olive navigates electronic medical records, logs into hospital portals, creates reports, files insurance claims, and more.
Olive does so thanks to three key traits. First, using computer vision and Robotic Process Automation, or RPA, the program can interact with any software interface just as a human would, opening browsers and typing. Second, machine learning enables Olive to make decisions the way human health care workers do. The team trained Olive with historical data on how health care workers perform digital tasks, such as how to file an insurance eligibility check for a patient seeking to undergo a procedure.
Tomi Engdahl says:
A new way to explain neural networks
https://techcrunch.com/2018/09/26/a-new-way-to-explain-neural-networks/?sr_share=facebook&utm_source=tcfbpage
By now, most of us have a general idea of what a neural network is, at least insomuch as its role in enabling the “machine learning” part of what’s considered AI today. Also known as deep learning
Explaining exactly how artificial neural networks (ANN) work in a mathless way can sometimes feel like a lost cause, though. They’re often likened to neural pathways in the human brain, but that’s not quite it, either, and the comparison is lost on anyone who didn’t pay attention in science class.
filmmaker Ben Sharony and PokeGravy Studios have done in A.N.N., an animated short
A.N.N. makes several mistakes, until, through trial and error (and feedback that nicely sums up the back-propagation process), she finally learns to identify (and find) the proper item.
Tomi Engdahl says:
COMPUTERS CAN SOLVE YOUR PROBLEM. YOU MAY NOT LIKE THE ANSWER.
https://apps.bostonglobe.com/ideas/graphics/2018/09/equity-machine/
What happened when Boston Public Schools tried for equity with an algorithm
Last year, the Boston Public Schools asked MIT graduate students Sébastien Martin and Arthur Delarue to build an algorithm that could do the enormously complicated work of changing start times at dozens of schools — and rerouting the hundreds of buses that serve them.
The algorithm was poised to put Boston on the leading edge of a digital transformation of government. In New York, officials were using a regression analysis tool to focus fire inspections on the most vulnerable buildings. And in Allegheny County, Pa., computers were churning through thousands of health, welfare, and criminal justice records to help identify children at risk of abuse.
The potential, says Stephen Goldsmith, a former mayor of Indianapolis who now runs the Data-Smart City Solutions project at Harvard University, is enormous: “more effective utilization of public resources, more individuals helped, more problems preempted.”
Dataphiles say algorithms may even allow us to filter out the human biases that run through our criminal justice, social service, and education systems. And the MIT algorithm offered a small window into that possibility.
Or, the whole thing could turn into a political disaster.
Tomi Engdahl says:
Researchers Use AI to Identify People via Seismic Signals Produced from Footsteps
https://blog.hackster.io/researchers-use-ai-to-identify-people-via-seismic-signals-produced-from-footsteps-da4d1f8121e8
Tomi Engdahl says:
Microsoft’s machine learning tools for developers get smarter
https://techcrunch.com/2018/09/24/microsofts-machine-learning-tools-for-developers-get-smarter/?utm_source=tcfbpage&sr_share=facebook
Tomi Engdahl says:
AI Could Provide Moment-by-Moment Nursing for a Hospital’s Sickest Patients
https://spectrum.ieee.org/biomedical/devices/ai-could-provide-momentbymoment-nursing-for-a-hospitals-sickest-patients
The ICU of the future will make far better use of its machines and the continuous streams of data they generate. Monitors won’t work in isolation, but instead will pool their information to present a comprehensive picture of the patient’s health to doctors. And that information will also flow to artificial intelligence (AI) systems, which will autonomously adjust equipment settings to keep the patient in optimal condition.
Tomi Engdahl says:
37 Reasons why your Neural Network is not working
https://www.kdnuggets.com/2017/08/37-reasons-neural-network-not-working.html
Over the course of many debugging sessions, I’ve compiled my experience along with the best ideas around in this handy list. I hope they would be useful to you.
Tomi Engdahl says:
https://www.google.com/amp/s/amp.economist.com/special-report/2016/06/25/from-not-working-to-neural-networking#ampf=undefined
Tomi Engdahl says:
Bots replacing office workers drive big valuations
https://techcrunch.com/2018/09/29/bots-replacing-office-workers-drive-big-valuations/?utm_source=tcfbpage&sr_share=facebook
A lot of people still get paid to sit in offices and do repetitive tasks. In recent years, however, employers have been pushing harder to find ways to outsource that work to machines.
Venture and growth investors are doing a lot to speed up the rise of these worker-bots.
So far this year, they’ve poured hundreds of millions into developers of robotic process automation technology, the term to describe software used for performing a series of tasks previously carried out by humans.
And employers are willing to pay handsomely to liberate their employees. UiPath said that in one 21-month period, it went from $1 million to $100 million in annual recurring revenue, an absolutely astounding growth rate for an enterprise software company.
software doesn’t eliminate jobs so much as it gives workers time to focus on higher-value projects.
Tomi Engdahl says:
Why Technology Favors Tyranny
https://www.theatlantic.com/magazine/archive/2018/10/yuval-noah-harari-technology-tyranny/568330/
Artificial intelligence could erase many practical advantages of democracy, and erode the ideals of liberty and equality. It will further concentrate power among a small elite if we don’t take steps to stop it.
Tomi Engdahl says:
Low-Cost, Low-Power AI on the Edge
https://www.eeweb.com/profile/max-maxfield/articles/low-cost-low-power-ai-on-the-edge
Expanded features in the Lattice sensAI stack are designed to speed time to market for developers of flexible machine-learning inferencing in consumer and industrial IoT applications
Tomi Engdahl says:
AI-powered IT security seems cool – until you clock miscreants wielding it too
Field both embraced, feared by enterprise
https://www.theregister.co.uk/2018/10/01/can_ai_be_trusted_on_security/
We’re hearing more about AI or machine learning being used in security, monitoring, and intrusion-detection systems. But what happens when AI turns bad?
Two interesting themes emerged from separate recent studies: the growth of artificial intelligence coupled with concerns about their potential impact on security.
Tomi Engdahl says:
Artistic Collaboration With AI
https://hackaday.com/2018/10/02/artistic-collaboration-with-ai/
Ever since Google’s Deep Dream results were made public several years ago, there has been major interest in the application of AI and neural network technologies to artistic endeavors. [Helena Sarin] has been experimenting in just this field, exploring the possibilities of collaborating with the ghost in the machine.
https://thegradient.pub/playing-a-game-of-ganstruction/
Tomi Engdahl says:
https://jyripaasonen.fi/voidaanko-tulevia-rikoksia-ennustaa/
Tomi Engdahl says:
Disrupt SF 2018
5 takeaways on the state of AI from Disrupt SF
https://techcrunch.com/2018/09/28/5-takeaways-on-the-state-of-ai-from-disrupt-sf/
Predictions of where the tech, money and morals are heading in artificial intelligence
Tomi Engdahl says:
Xilinx Details SoC-like FPGAs
Versal aims to rival Intel, Nvidia processors
https://www.eetimes.com/document.asp?doc_id=1333815
Xilinx released the first details of its next-generation Everest architecture, now called Versal. It shows the microprocessor landscape is blurring as CPUs, GPUs and FPGAs morph into increasingly similar SoC-like devices.
Versal shrinks the size of a central FPGA block to make room for more ARM, DSP, inference and I/O blocks. It comes as Intel and AMD make room for beefier GPUs in their x86 chips and Nvidia adds specialty cores for jobs like deep learning on its GPUs.
Xilinx positioned Versal as the start of a broad new family of standard products. They aim to outperform CPUs and GPUs on a wide range of data center, telecom, automotive and edge applications and increasingly support programming in high-level languages such as C and Python.
Tomi Engdahl says:
MIT researchers develop new chip design to take us closer to computers that work like human brains
https://www.cnbc.com/2018/10/08/mit-develops-a-chip-to-help-computers-work-more-like-human-brains-.html
Scientists at MIT are developing brains-on-a-chip for neuromorphic computing.
It would allow processing facts, patterns and learning at lightning speed and could fast-forward the development of humanoids and autonomous driving technology.
Last year the market for chips that enable machine learning was approximately worth $4.5 billion, according to Intersect360.
Tomi Engdahl says:
Google’s Pixel 3 will use AI to respond to telemarketer calls
https://thenextweb.com/plugged/2018/10/09/googles-pixel-3-will-use-ai-to-respond-to-telemarketer-calls/
Google today announced its Duplex AI feature will roll out to Pixel 3 owners next month, but that’s not the only cool new AI experience they’ll get. A new call-screening and response feature will launch with Duplex, and it could mean an end to spam calls. Telemarketers: Your days are numbered.
Duplex is the futuristic jaw-dropping AI the company showed off earlier this year that calls businesses to schedule appointments for you autonomously. It’s finally making its way to English language customers sometime next month and, along with it, a robust new update to Google’s call screening AI is slated for debut as well.
Tomi Engdahl says:
Sarah Perez / TechCrunch:
Google launches Live Albums, which uses AI to automate the sharing of photos with groups of people and displays them on the Home Hub as they are taken — Just ahead of today’s Google’s hardware event, the company quietly rolled out an update to its Google Photos application which introduces a new …
Google Photos adds automated sharing through ‘Live Albums,’ which can stream to Home Hub
https://techcrunch.com/2018/10/09/google-photos-adds-live-albums-an-automated-way-of-sharing-people-pet-photos-with-anyone/
Tomi Engdahl says:
Josh Constine / TechCrunch:
Instagram will use machine learning to scan photos to detect bullying and send them to community moderators for review if needed, rolling out in coming weeks — Instagram and its users do benefit from the app’s ownership by Facebook, which invests tons in new artificial intelligence technologies.
Instagram now uses machine learning to detect bullying within photos
https://techcrunch.com/2018/10/09/instagram-bullying-photos/
Instagram and its users do benefit from the app’s ownership by Facebook, which invests tons in new artificial intelligence technologies. Now that AI could help keep Instagram more tolerable for humans. Today Instagram announced a new set of antii-cyberbullying features. Most importantly, it can now use machine learning to optically scan photos posted to the app to detect bullying and send the post to Instagram’s community moderators for review. That means harassers won’t be able to just scrawl out threatening or defamatory notes and then post a photo of them to bypass Instagram’s text filters for bullying.
Tomi Engdahl says:
Jeffrey Dastin / Reuters:
Sources detail how Amazon shut down a machine learning tool for rating job applications in 2017 because it was biased against female candidates — SAN FRANCISCO (Reuters) – Amazon.com Inc’s (AMZN.O) machine-learning specialists uncovered a big problem: their new recruiting engine did not like women.
Amazon scraps secret AI recruiting tool that showed bias against women
https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G
Amazon.com Inc’s (AMZN.O) machine-learning specialists uncovered a big problem: their new recruiting engine did not like women.
The team had been building computer programs since 2014 to review job applicants’ resumes with the aim of mechanizing the search for top talent
Automation has been key to Amazon’s e-commerce dominance, be it inside warehouses or driving pricing decisions.
“Everyone wanted this holy grail,” one of the people said. “They literally wanted it to be an engine where I’m going to give you 100 resumes, it will spit out the top five, and we’ll hire those.”
Tomi Engdahl says:
Frederic Lardinois / TechCrunch:
Nvidia launches Rapids, an open source platform for data science and machine learning, to help bring GPU acceleration to data analytics, with IBM, HPE, others — Nvidia, together with partners like IBM, HPE, Oracle, Databricks and others, is launching a new open-source platform for data science and machine learning today.
Nvidia launches Rapids to help bring GPU acceleration to data analytics
https://techcrunch.com/2018/10/10/nvidia-launches-rapids-to-help-bring-gpu-acceleration-to-data-analytics/
Tomi Engdahl says:
Deep learning? Here’s how to exercise your neural networks
Getting practical with machine learning and AI
https://www.theregister.co.uk/2018/10/11/deep_learning_heres_how_to_exercise_your_neural_networks/
Tomi Engdahl says:
Making AI Run Faster
https://semiengineering.com/making-ai-run-faster/
Why inferencing is the next battleground.
The semiconductor industry has woken up to the fact that heterogeneous computing is the way forward and that inferencing will require more than a GPU or a CPU. The numbers being bandied about by the 30 or so companies working on this problem are 100X improvements in performance.
But how to get there isn’t so simple. It requires four major changes, as well as some other architectural shifts.
Tomi Engdahl says:
https://www.tivi.fi/Kumppaniblogit/dna/algoritmit-mellastavat-pian-pilvessa-pahat-mielessa-6742440
Weaponized drones. Machines that attack on their own. ‘That day is going to come’
https://www.cnbc.com/2018/07/20/ai-cyberattacks-artificial-intelligence-threatens-cybersecurity.html
Artificial intelligence has clear positive uses, but it could be used to teach machines to attack people and their computer networks on their own.
Drones and autonomous vehicles could be hacked using AI and turned into weapons
Traditional cybersecurity methods won’t know how to cope with new attacks carried out by smart machines.
Tomi Engdahl says:
DARPA wants to teach and test ‘common sense’ for AI
https://techcrunch.com/2018/10/11/darpa-wants-to-teach-and-test-common-sense-for-ai/?sr_share=facebook&utm_source=tcfbpage
AdChoices
DARPA wants to teach and test ‘common sense’ for AI
Devin Coldewey
@techcrunch / 23 hours ago
Artificial intelligence and cybernetics
It’s a funny thing, AI. It can identify objects in a fraction of a second, imitate the human voice and recommend new music, but most machine “intelligence” lacks the most basic understanding of everyday objects and actions — in other words, common sense. DARPA is teaming up with the Seattle-based Allen Institute for Artificial Intelligence to see about changing that.
The Machine Common Sense program aims to both define the problem and engender progress on it, though no one is expecting this to be “solved” in a year or two.
https://www.darpa.mil/news-events/2018-10-11
Tomi Engdahl says:
Google AI claims 99% accuracy in metastatic breast cancer detection
https://venturebeat.com/2018/10/12/google-ai-claims-99-accuracy-in-metastatic-breast-cancer-detection/
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.
Tomi Engdahl says:
https://pulse.microsoft.com/fi-fi/business-leadership-fi-fi/na/fa1-perheenisasta-potilaaseen-tekoaly-koskettaa-kaikkialla/?utm_campaign=AI&utm_medium=content_social_ad60_fifi&utm_content=none_FY19_Q2_BDM&utm_source=facebook&ocid=AID_oo_spl100000334889928
Tomi Engdahl says:
Amazon’s Secret AI Hiring Tool Reportedly ‘Penalized’ Resumes With the Word ‘Women’s’
https://gizmodo.com/amazons-secret-ai-hiring-tool-reportedly-penalized-resu-1829649346?utm_medium=sharefromsite&utm_source=Gizmodo_facebook&utm_campaign=sharebar
For years, a team at Amazon reportedly worked on software that vetted the resumes of job applicants in an effort to surface the most likely hires. It gradually became clear that no matter how hard engineers tried to fix it, the recruitment engine found a way to discriminate against women, Reuters reports.
The goal was to teach an AI how to identify the most likely hires to streamline the list of potential recruits that would have to be subsequently vetted by human recruiters.
In effect, Amazon’s system taught itself that male candidates were preferable. It penalized resumes that included the word “women’s,” as in “women’s chess club captain.” And it downgraded graduates of two all-women’s colleges
Gizmodo reached out to Amazon for comment on the report and a spokesperson sent us the following statement: “This was never used by Amazon recruiters to evaluate candidates.”
The algorithm’s gender discrimination issues became apparent about a year into the project’s lifecycle and it was eventually abandoned last year
the team just couldn’t get the tool to stop reverting to discriminatory practices, Reuters reported. As time went on, the models often spiraled into recommending unqualified applicants at random.
The team’s effort highlights the limitations of algorithms as well as the difficulty of automating practices in a changing world.
Tomi Engdahl says:
Mack DeGeurin / New York Magazine:
How facial recognition is being used to monitor animals: detecting sickness in salmon, gauging diet of cattle, protecting elephants from poachers, and more — Facial recognition technology has some serious, persistent issues. These were clearly shown earlier this year when Amazon’s …
Here Is a List of Every Animal Humans Currently Monitor Using Facial Recognition Technology
http://nymag.com/developing/2018/10/what-creatures-may-we-place-in-the-panopticon.html
Tomi Engdahl says:
How Machines Learn
https://www.youtube.com/watch?v=R9OHn5ZF4Uo
How do all the algorithms around us learn to do their jobs?
making the bots look adorable and harmless is how we end up with skynet
Tomi Engdahl says:
Making AI Run Faster
Why inferencing is the next battleground.
https://semiengineering.com/making-ai-run-faster/
The semiconductor industry has woken up to the fact that heterogeneous computing is the way forward and that inferencing will require more than a GPU or a CPU. The numbers being bandied about by the 30 or so companies working on this problem are 100X improvements in performance.
Tomi Engdahl says:
https://www.tivi.fi/Kaikki_uutiset/suomi-voi-edeta-tekoalyn-karkimaaksi-tarvitaan-kuitenkin-asennemuutos-ja-runsaasti-tilastotieteen-opiskelua-6745002
Mikä näyttäisi olevan vaikeinta tekoälyn käyttöönotossa?
”Asenteiden muutos on suurempi haaste kuin tekniikan muutos. Tekoälystä hyötyminen edellyttää, että tuotekehitys, valmistus, myynti ja huolto toimivat saman datan perusteella. Toki osa haasteista on teknisiä ja liittyy esimerkiksi datan käytettävyyteen, validointiin ja algoritmien rakentamiseen.”
Miten tärkeää on, että yritys kehittää oman sovelluksen datan analysointiin?
”Tarjolla on kaupallisia ratkaisuja, siis laitteita datan keräämiseen ja pilvipalveluja sekä laskentatehoa ja sovelluksia datan käsittelyyn. Tekoäly vaatii kuitenkin ohjelmistokehityksen sijaan ennen kaikkea analyytikkoja, jotka pystyvät kertomaan datan merkityksen. Tekoälyn käytön lisääntyessä ohjelmistokoodin voi ostaa, mutta omia analyytikoita tarvitaan lisää. Cargotec luokitellaan konepajaksi. Meillä on kuitenkin töissä sata kertaa enemmän analyytikoita ja koodaajia kuin hitsaajia.”