News — At The Edge — 7/28

• Born under a bad sign — AI learning, trolling, economy & political influence — leads to unwanted developments and consequences.
• Shape of things to come — health privacy, insect bots & chip making — is a mixed bag of developments.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Born under a bad sign -

“[DARPA] is looking to change the way works…[with] …that learn continuously, adapt to new tasks, and know what to learn and when. ‘

We want the rigor of automation with the flexibility of the human’….

Neural networks are adaptable systems whose ability to learn comes from…[training] on a set of data….[Problem] when the AI encounters something it was never trained to recognize….[So] two sets of groups.

One set will…develop systems that can continuously learn, adapt to new tasks and circumstances, and understand inputs according to what the system’s mission is (called ‘goal-driven perception)….

[Another] to identify new mechanisms of lifelong learning — from biology or a physical science — and transfer that mechanism to an algorithm that improves AI.”

“[T]he use by states of targeted online hate and harassment campaigns to intimidate and silence individuals critical of the state…[using] surveillance and hacking…afforded by a new era of pervasive technology…persecutes perceived opponents at scale….

[There’s] several commonalities…[including] death and rape threats, using bots…to amplify vitriolic attacks at scale…accusations of treason or collusion with foreign intelligence agencies, using ‘black’ public relations firms to disseminate hyper-partisan or libelous disinformation about targets, spreading doctored images and memes…using spyware and hacking to gather…intelligence against targets, and sowing acrimonious sexism….

Changes in law are unlikely to effectively stem the practice of state-sponsored trolling in the short term…[so] technology companies bear…[responsibility] to curb…state-sponsored harassment campaigns…[as] form of human rights abuse…with pinpoint personalization….[Need]

• the expansion of…international and US law to include the types of online harassment and hate speech deployed as part of state-sponsored trolling attacks;
• the reconsideration of intermediary liability…[for] platforms in facilitating trolling campaigns;…
• technology companies to detect and identify state-linked accounts, bots, and hateful content online.”

“[Prosperity] constructs a society with greater economic and social justice…[Trump’s] appeal rests heavily on our loss of confidence in this vision….

From 1950 to 1973, the economy grew at an average annual rate of …[but] from 2008 to 2017…averaged only 1.5 percent….[T]he postwar boom…[had] three economic advantages…that were fated to fade….

1. backlog of new technologies (television, jet aircraft, synthetic fibers, antibiotics, air conditioning) that boosted both consumer spending and business investment….
2. destruction of Europe and Japan left…few serious [competitors]…[so] trade balance was routinely in surplus; the [1971]….
3. progress in stabilizing the economy….2008–2009 financial crisis remind us that the problem of the business cycle hasn’t been solved….

U.S. corporations turn out to be vulnerable…abroad, and to new technologies…. The skewing of income toward those at the top compounds the effect…[and] a breeding ground for discontent….

Trump blames… and imports…[and] American officials…[so] tariff proposals seem to be causing as much or more grief among the U.S. firms they’re supposed to help…[because] there’s a big difference between economic nostalgia and economic policy.”

“SQUEEZING the top 1% ought to be…[easy] for politicians seeking to please the masses…[but] more concerned with immigration and sovereignty…[is] sign of the corrupting influence of inequality….

19th century governments across the West faced the threat of socialist revolution…[and] promises of greater redistribution…[needed] institutional guarantees….

America’s electoral college…prized by elites…because they served as a check on the egalitarian tendencies of the masses….

[Today] more than two-thirds of Americans and Europeans express concern about current levels of inequality…[but] democracies might have become too jammed to make progress on any issue of substance….

Last year America’s government…change to taxes…tilts the distribution of income even more in favor of the rich…[and] politicians seem less interested in grappling with the problem….

[So] rising inequality…boost the power of the rich…to counter the popular will…[with] political preferences of those with $40m or more…overwhelmingly favor cutting spending on major social-safety-net programs. (The general public wants it increased.).…

[In] regular personal contact with elected officials…fewer than 30,000 people account for a quarter of all national political donations from individuals and…more than 80% of the money raised by political parties…[with] means to shape public opinion: financing…think-tanks…or buying media outlets…to shape public narratives about which problems deserve attention…[so] economic justice are crowded out….

[It’s] the ‘negative agenda power’ of the rich…that reducing inequality becomes less likely even as it becomes more urgent…with a loss of democratic accountability as a nasty side-effect….

[So far] across human history, inequality inevitably rises until checked by disasters like wars or revolutions….If political leaders tried…they might…find that redistribution is a winner at the ballot box.”

Shape of things to come —

“[Insurers] are tracking your race, education level, TV habits, marital status, net worth**…social media, whether you’re behind on your bills, what you order online…[then] feed this information into…algorithms that spit out predictions about how much your health care could cost them….**

[But] patient advocates and privacy scholars say the insurance industry’s data gathering runs counter to its…federally required, allegiance to patients’ medical privacy….

HIPAA only protects medical information….Patient advocates warn that using unverified, error-prone ‘lifestyle’ data to make medical assumptions could lead insurers to improperly price plans — for instance, raising rates based on false information — or discriminate against anyone tagged as high cost….

[This] is important because patients — through taxes, cash payments and insurance premiums…[fund] entire health care system. Yet the industry’s…strategies and inside deals often has little to do with patients’ needs….

[T]he big guns in health data…[are] Optum, IBM Watson Health and LexisNexis….[Optum] owned by…UnitedHealth Group…collected the medical diagnoses, tests, prescriptions, costs and socioeconomic data of 150 million Americans going back to 1993…to link patients’ medical outcomes and costs to details like their level of education, net worth, family structure and race…[and] filed a patent…to gather what people share on…Facebook and Twitter, and to link this…to the person’s clinical and payment information….

Doctors and hospitals have typically been paid based on the quantity of care [provided]…[but] moving toward paying them in lump sums for caring for a patient, or for an event, like a knee surgery…[so] can profit more when patients stay healthy….

Studies show social and economic aspects of people’s lives play an important role in [health]…[but] industry has a history of boosting profits by signing up healthy people and finding ways to avoid sick people….

Affordable Care Act prohibits insurers from denying people coverage based on pre-existing health conditions or charging sick people more…[but] Trump administration is promoting short-term health plans, which do allow insurers to deny coverage to sick patients…[fail] to include enough information about which drugs a plan covers…[change] things a plan covers, or how much a patient has to pay…after a patient has enrolled…[or] exclude or limit certain types of providers from their networks — like those who have skill caring for patients with HIV or hepatitis C….

[IBM] surveys 80,000 Americans a year to assess lifestyle, attitudes and behaviors that could relate to health care…[and] analyzed together to identify social and economic factors for an area….A region with too many sick people, or with patients who don’t take care of themselves, might not be worth the risk….’God forbid you live on the wrong street…[and] get lumped in with a lot of bad things’….

[LexisNexi] uses 442 nonmedical personal attributes to predict a person’s medical costs…[with] more than 78 billion records…including people’s cellphone numbers, criminal records, bankruptcies, property records, neighborhood safety and more…to predict patients’ health risks and costs in eight areas….

People who downsize their homes tend to have higher health care costs…[likewise] those whose parents didn’t finish high school. Patients who own more valuable homes are less likely to land back in the hospital within 30 days of their discharge….[T]he largest health actuarial firm in the world, Milliman…[uses] LexisNexis scores…[for] individual patients and make adjustments to protect themselves from losses….

’No one gave anyone permission to do this…[and] not health data…[but] inferred data’…[leading] to poor people being charged more…[and] harder for them to get the care they need….[Employers] could even decide not to hire people with…high medical costs in the future….

[Our] policymakers could…protect people’s information…[like] ’Europe, data protection is a constitutional right’….[Also] health scores should be treated like credit scores. Federal law gives people the right to know their credit scores and how they’re calculated….[If] rated by whether they listen to sad songs…or look up information about AIDS online, they should know….

[LexisNexis] told me that if it had calculated my scores it would have done so on behalf of…my insurance company. So, I couldn’t have them.”

“Companies are waging an invisible data war online…[using] software to scan rival websites and collect prices, a process called ‘scraping’…[with] machine-learning algorithms to help…decide how much to charge for different products….

Google and Bing scrape web pages…for their search engines. Academics and journalists use scraping software to gather data….

[But] scraping can be a two-way street…[with] defenses to subvert scraping….

[One’s] showing different prices to real people than to ….[So some] mask bots to avoid detection…[or] appear to be coming from smartphones….[Users] choice: View ads or allow the app to use ‘some of your device’s resources….Sometimes bots actually tell the sites they’re visiting…[but] can also lie. One technique for detecting bots is the frequency with which a visitor hits a site….Captchas can help, but…an inconvenience for legitimate users….

[Trouble’s] is the need to allow some, but not all, bots to scrape a site. If websites blocked bots entirely, they wouldn’t show up in search results. Retailers also generally want their pricing and items to appear on shopping comparison sites….

[Amazon] ‘prioritize humans over bots’….[Publishers] make sure advertisers are showing a site’s viewers the same ads that they show to the publishers….[Recently] Chrome extension was of being used to steal passwords.”

“[Large robots] capable in some very specific situations, but…smaller, cheaper, more specialized robots is much more efficient…[especially] with data collection as opposed to manipulation….

[DARPA] program called SHRIMP: SHort-Range Independent Microrobotic Platforms…goal is ‘to develop and demonstrate multi-functional micro-to-milli robotic platforms….

[Tech] advances in microelectromechanical systems (MEMS), additive manufacturing, piezoelectric actuators, and low-power sensors have allowed researchers to…to create the strength, dexterity, and independence of functional microrobotics platforms….

DARPA expects…systems that weigh less than one gram and fit into one cubic centimeter….[to]

• High Jump…vertically from a stationary starting position…and survivability…>5cm.
• Long Jump…horizontally from a stationary starting position…and survivability…>5cm
• Weightlifting…progressively larger masses until…>10g.
• Shotput_…_>10cm @ 1g, >5cm @ 2g.
• Tug of War…connected to a load…> 25mN…
• Rock Piling…travel to, lift, and stack weights…in a minimum of two layers without human interaction…2g, 2 layers.
• Steeplechase…traverse the course without human interaction or recharge between each obstacle…2 obstacles, 5m.
• Biathlon…traverse to a series of beacon waypoints to create an open circuit without human interaction or recharge between each waypoint…2 beacons, 5m.
• Vertical Ascent…traverse up two surfaces, one with a shallow incline (10º) and the other with a sharp incline (80º)…10m at 10°, 1m at 80°.”

—

“[T]o make small groups of engineers capable of feats that would take 100 engineers to achieve today…aiming for 2025 to 2030… around chip architecture…of deep neural nets and other implementations**….**

[A] program called …[is] making associations across data sets which aren’t densely connected…doing things like updating memory access patterns, updating the type of cores that would be doing the processing, and working across the software stack to do a hardware/software codesign for a variety of applications….

[Next] is something we call ‘software-defined hardware’…[where] hardware is smart enough to reconfigure itself to be the type of hardware you want, based on an analysis of the data type that you’re working on….[The] hard thing is…how to do that data introspection, how to reconfigure the chip on a…millisecond timescale…monitor whether you’re right…[and] constantly evolving toward the ideal solution….

[Lastly] is the ‘domain specific system-on-chip…starting with the software-defined radio…[then] to machine vision and machine learning and other domains to see if you can still have simplified programming models running on top of hyperspecialized hardware….[Work] showed that two to three student designers can do full system-on-chip designs by abstraction and automation….

Idea is really the intersection of machine learning and electronic design automation (EDA)…to capture the capabilities of the designer inside the EDA itself…[so] every time you use an EDA package, it gets smarter and…next design [easier]…envision a cloud resource where multiple people are sharing and can all get better simultaneously….

[Today] rare to go to GitHub, find high-quality hardware blocks that are available, and have the verification tools and everything you would need to trust that…it is in a state that is useable for your design. Posh is…about the verification tools…[and] IP blocks that will be freely available….

As the abstraction of the hardware design gets higher and higher, it gets closer to the software community’s mentality…[and] a methodology to understand how good something is at a deep level before it’s used….

[Also] look at where memory is stored and ask: Can those same materials be used for processing information?…. Next-generation nonvolatile memory, for example, that and with a low write energy.”

Find more of my ideas on Medium at,

.

Or click the Follow button below to add me to your feed.

Prefer a weekly email newsletter — free, no ads, no spam, & never sell the list — email me with “add me” in the subject line.

May you live long and prosper!Doc Huston