Rebane's Ruminations

Hockeystick Hansen Strikes Again (updated 7aug12)

"In fact, [climate change] has now driven our climate outside the range that has existed the last 10,000 years..." -- Dr. James Hansen (from NPR)

George Rebane

There’s a coordinated blitz to recover ground lost to climate change skeptics in the past couple of years.  The lost ground is uneven – in fact, in California there’s actually been a gain as the rogue CARB continues creating chaos – but there’s been enough doubt expressed by politicians across the country so that another application of bogus science is called for to shore up the politically correct belief systems that have recently shown some fraying at the edges.

And who but that intrepid NASA scientist Hockeystick Hansen himself has leaped into the very breach of the breech with a Washington Post op-ed piece, and a report purported to be published by the National Academy of Sciences (I couldn’t find it, and their search engine has no knowledge of a James Hansen.  Hmmm.).  But there’s enough in the political coverage (here and here) of ol’ Hockeystick’s re-emergence to piece together the elements of his current assault on the nation’s credibility.

RR readers have been exposed to his shenanigans in past years when he first introduced his notorious global temperature ‘hockeystick’ (here and here).  And, in addition to scores of scientists, tireless bloggers like Anthony Watts, Steve McIntyre, and our own Russ Steele have spent years laying bare the errors and lack of science that the AGW (anthropogenic global warming) crowd assembled by the UN has been inflicting on the world’s public policies.

Well, now Dr Hansen claims that all these extreme hot weather events we’ve had in the last few years are proof positive that they are caused by AGW.  BTW, the cold weather events during the same interval are just that, ‘weather’ and not climate change.  You gotta have a PhD to tell the difference.

And this time he’s not bringing up another collection of dodgy computer models like those that created the hockeystick.  No, now he’s appealing to established statistics and recorded historical weather data – you know, bell curves and all that.  Well actually it’s hard to tell what kind of statistics he’s using because the media are guaranteed to muck up what he really did.  All we know from the reports is that he compared data from the base period of 1951-1980 with data from 1981-2010 – two thirty-year periods from a (climate) process that changes in the order of centuries.  Best to watch the NPR video to get the gist of his arguments.

Assuming he carefully tallied up weather events of various intensities for the base period, he could then make a histogram that plots weather intensity (x-axis) against the number of such incidents for each level of weather intensity (y-axis).  He then fits a bell curve (Gaussian distribution) to this histogram.  Then he does the same thing for the 1981-2010 interval data, and lays the two bell curves over each other.  And by Jove, it appears that the most recent bell curve is visibly shifted to the right of the base period bell curve.

The extreme weather events are those represented by the right tails of the two bell curves (see Hansen’s figures in the video), and the most recent 30 years shows a higher number for any intensity level of weather events.  So there you have it, slam dunk, end of story, AGW is here, let’s get that cap n’tax legislation going again.

Judea Pearl, Turing Laureate

George Rebane

Judea Pearl, Professor of Computer Sciences UCLA, received the A.M. Turing Award last night at the awards banquet of the Association of Computing Machinery at the Palace Hotel, San Francisco.  He now joins the forty-three previous laureates of the ‘Nobel prize of computing’, of whom twenty-five were in attendance to honor him.

Jo Ann and I joined the gala celebration as guests of the awardee, and sat with family and friends, most of whom we have known for decades.  (I am a lifelong member of the ACM.)  The awards banquet also honored other contributors to the field and culminated in the Turing Award presentation.  Here is ACM’s official, yet still modest, recounting of Judea’s accomplishments.

The Pearls became family friends soon after Judea Pearl collaborated on a project with a ‘black studies’ company I was running in the 1970s.  Judea’s wife Ruth (an accomplished software engineer in her own right) is a consummate hostess.  Jo Ann also had (still has) a reputation for putting on mean soirees that, with the passing hours, devolved into impromptu performances ranging from poetry recitations, through heated debates on various arcane and absorbing topics, to planned revivals of long ago salons of the Englightment.

We were often at each other’s house for such dinners, parties, and special noodling sessions.  These always included music that had us singing with guitars and keyboard for accompaniment.  Judea is an accomplished guitar player, keyboardist, and has a great singing voice.  He’s also a world class choral conductor and has performed numerous times at the LA Music Center.  (I was the less practiced one on both instruments, making up in volume for what I lacked in technique.)

A couple of shared memories from a rich catalog.  For Ruth’s birthday one year Judea invited us and another couple to a small westside bistro in LA.  We knew that something special was going to happen because Judea brought his guitar; but what could that be since we were eating in a restaurant?  Not a problem, soon the main course was finished, and after dinner drinks were in hand.  The lights were low and the atmosphere was perfect as Judea reached behind his chair for his trusty guitar and putting a foot on his chair, he launched into a medley of love songs to his bride of many years.

The other patrons in the restaurant were enthralled along with those at our table.  I recall Judea ending his special gift to Ruth (and the rest of us) with a rendition of a soulful Spanish adagio for solo trumpet and guitar from a piece I could not place.  It was a night to remember, as Judea kissed Ruth to the applause of the entire restaurant.  (Solo trumpet?  Did I fail to mention that Judea also has a unique gift, the ability to purse his lips, puff his cheeks, and perfectly simulate the sound of a virtuoso trumpet?)

Another memory is when I got a call late one afternoon from Judea saying that Ed Feigenbaum was in town and could drop by for some dinner and talk; would I care to join them?  (The techie reader will recognize Feigenbaum as a renown giant of AI, ‘father of expert systems’, and Turing laureate who taught for years at Stanford - now professor emeritus.)  Long story short.  After one of Ruth’s ‘oh it’s nothing’ fabulous meals, the three of us retired to the patio and spent the next four (five?) hours talking about the state of machine intelligence, who was doing what, and possible directions the science would take and its impact on society.  At the time Judea had already published his landmark 1986 paper and was working on his 1988 book on Bayes nets (I was one of Judea’s doctoral students doing research in Bayes learning).  It was clear that the torch would soon be passed from Feigenbaum’s expert systems to Pearl’s Bayes nets (with causality soon making its early murmurs).  Truly another memorable night.
   
So that was the basis for all the fuss this weekend at the Rebane household.   Yesterday morning we dropped off our puppy with the Steeles, and headed through a hot and crowded central valley to San Francisco.  Judea and I were able to spend some quality time together before the band started playing.  (The picture with the other white haired professor catches a debate on who has known Judea the longest - I foolishly thought c'est moi.  I came in a distant second to the good professor who has known Judea since they were playmates on an Israeli kibbutz eons ago while the Brits still ran Palestine.)

Jo Ann and Ruth picked up where they left off (top picture).  Ruth also sits on the board of the Daniel Pearl Foundation which is dedicated in memory of their son and has as its mission statement –

The Daniel Pearl Foundation has been formed by Danny's family and friends to continue Danny's mission and to address the root causes of this tragedy, in the spirit, style, and principles that shaped Danny's work and character. These principles include uncompromised objectivity and integrity; insightful and unconventional perspective; tolerance and respect for people of all cultures; unshaken belief in the effectiveness of education and communication; and the love of music, humor, and friendship.

(Full disclosure: Jo Ann and I are ongoing supporters of the foundation, especially its work in bringing greater mutual understanding between Islamic and western cultures.)

The Bayes Surface and more (updated 19jan13)

 George Rebane

[This post continues the numeracy series on the Bayes contribution to our civilization.  For more background and examples, please see ‘Making Medical Decisions’.]   


What you see here is a 3D graphical illustration of the celebrated Bayes Theorem (or formula).  I have not seen this beautiful surface presented in the literature, so I wrote a little program to do just that.  In this representation the Bayes formula (seen above the surface) is expressed in its likelihood form.  We are reminded that the Bayes Theorem is the correct and most powerful process by which we take existing uncertain knowledge, combine it with new data or observations, and update that knowledge.

Today the Bayes Theorem underlies the operational principles embedded in so much of what makes our modern world.  It is safe to say that most new technologies in fields that include medicine, energy exploration and production, finance, communication, space exploration, transportation (land, sea, air), defense, the internet, manufacturing, distribution of goods and power, … would be handicapped beyond belief were we to remove their Bayesian components – we would literally be thrown back into the technological age of the 1950s, and our modern world would cease to exist.

To the above list we add artificial or machine intelligence (AI for short), all forms of which are now based on Bayesian inference and its newest offspring, the new calculus of causality.  These applications focus on learning from the real world, a world which presents itself ambiguously within confounding frames of uncertainty.   As critters, our brainbones have evolved to make Bayesian kinds of decisions within a utility that ranks survival highest.  Many of our neural structures appear to have parts of the illustrated surface hardwired into what Daniel Kahneman (Thinking Fast and Slow, 2012) has identified as our intuitive ‘System 1’ cognitive processor that lets us come to rapid conclusions.

Courtroom Evidence – the Z/M case (updated 28apr12)

George Rebane

Reason and logic have always had a low survival rate in courtrooms, especially where poorly understood evidence of a technical nature is presented by self-serving lawyers to innumerate juries.  In recent memory, the OJ Simpson murder trial may serve as a nationally prominent Exhibit A for this proposition.

Today we have the much examined and touted (for various agendas) Zimmerman/Martin shooting case that will soon be going to trial.   Much effort and focus is put on a piece of recorded audio that appears to have become the evidence by which Zimmerman’s claim of justifiable homicide will stand or fall.  This audio clip is now undergoing forensic analysis by state agencies and private parties having court-acknowledged expertise in matters pertaining to analyzing speech.  Some of these experts have apparently weighed in with some “preliminary findings” which now serve as grist for the media mill and a large gaggle of very opinionated bloggers, commenters, and self-appointed jurists in the blogosphere.

RR has chosen not to join this cohort opining on Zimmerman’s guilt or innocence, instead we wish to use the case to examine the larger questions of ‘stand your ground’ victims’ rights, and in this  piece how to properly look at reported evidence in such cases.  In the following we take a look at a report on the blog of “legal scholar” Professor Jonathan Turley by guest poster Mark Esposito.

I have no idea of Mr Esposito’s qualifications, but simply post his opening paragraph that also happens to be his summa.

The  chair emeritus for the American Board of Recorded Evidence, Tom Owen, and Ed Primeau, a Michigan-based audio engineer and forensics expert, have independently concluded that the furtive pleas for help clearly heard on the 911 tapes are not George Zimmerman’s. Both acknowledged experts used voice enhancing software, but different techniques, to rate the probability of the voice being Zimmerman’s at no more than 48%.  A 90% match is considered scientifically reliable.

So there you have it, the case is seemingly closed.  But not quite.

Making Medical Decisions

George Rebane

[This introduction to modern medical decision making is intended as the first of a series of such essays on RR.  Subsequent pieces will cover simplifying nomograms (to avoid the arithmetic), personal utility (always subjective), and planning sequential diagnostics and treatments.  The aim here is to prepare the intelligent healthcare customer to communicate effectively with the healthcare industry and make prudent decisions for his own benefit.  A related background tutorial by the author for this piece is 'Hello Bayes Theorem!']

The healthcare industry woefully lags in many areas, but helping patients make correct decisions about their individual healthcare is where healthcare providers, led by attending physicians, are pathetically behind the times.  Let me explain.

Judea Pearl wins 2011 Turing Award

George Rebane

The Association for Computing Machinery announced today that it has named Dr Judea Pearl of the University of California, Los Angeles as the recipient of the A.M. Turing Award for 2011.  The announcement itself begins –

NEW YORK, March 15, 2012—ACM, the Association for Computing Machinery today named Judea Pearl of the University of California, Los Angeles the winner of the 2011 ACM A.M. Turing Award for innovations that enabled remarkable advances in the partnership between humans and machines that is the foundation of Artificial Intelligence (AI). Pearl pioneered developments in probabilistic and causal reasoning and their application to a broad range of problems and challenges. He created a computational foundation for processing information under uncertainty, a core problem faced by intelligent systems. He also developed graphical methods and symbolic calculus that enable machines to reason about actions and observations, and to assess cause-effect relationships from empirical findings. His work serves as the standard method for handling uncertainty in computer systems, with applications ranging from medical diagnosis, homeland security and genetic counseling to natural language understanding and mapping gene expression data. His influence extends beyond artificial intelligence and even computer science, to human reasoning and the philosophy of science.

The Turing Award, widely considered the "Nobel Prize in Computing," carries a $250,000 prize, with financial support provided by Intel Corporation and Google Inc. It is named for the British mathematician Alan M. Turing, whose 100th anniversary will be celebrated in June at the ACM 2012 Turing Centenary Celebration that includes 34 past Turing Award winners along with Pearl.

"Like Alan Turing himself, Pearl turned his thinking to constructing procedures that might be harnessed to perform tasks traditionally associated with human intelligence," said Vint Cerf, chair of the ACM 2012 Turing Centenary Celebration, and a former ACM Turing Award recipient. "His accomplishments over the last 30 years have provided the theoretical basis for progress in artificial intelligence and led to extraordinary achievements in machine learning, and they have redefined the term 'thinking machine.'" Cerf pointed to Pearl’s innovation as a quantum leap from Turing’s "test" dating to the 1950s, when Turing set out to discover if machines could think. "Pearl's work on reasoning with uncertainty as well as his game˗changing contributions to machine reasoning about causality have had a pervasive influence not only on machine learning but on natural language processing, computer vision, robotics, computational biology, econometrics, cognitive science, and statistics," Cerf said.

The rest of the announcement may be read here.

Judea Pearl and I first became acquainted as colleagues on various defense research projects during the 1970s when he consulted for the small ‘black studies’ company that I headed during that period.  Subsequently our families became friends and over the years we spent many evenings in exciting discussions and song (Judea on guitar and me on the keyboard - Judea is an excellent musician in voice, choral, and several instruments.).  It was Judea who convinced me to complete my doctorate and join his small troupe of students at UCLA working on the various research tasks that served as the halo around the magnificent tour de force that Judea Pearl presented to the world when he showed how Bayesian networks could be constructed and correctly computed.

Once his 1988 Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference was published, Bayes nets became and have remained the state-of-the-art in machine intelligence.  Bayesian inference now is the standard in all applications that required correct probabilistic reasoning, decisions under uncertainty, and management of risk - in short in all human endeavors.  Every field that has not embraced what Pearl started with Bayes is now scrambling to catch up.

I have long had the honor of calling Judea Pearl my teacher, mentor, colleague, and friend.  Those of us who had the good fortune to recognize early his genius and accomplishments were amply rewarded by what he taught us, and how he guided our individual contributions (my own humble efforts were in machine learning and causality).  It is most appropriate that Judea now joins the pantheon of the recognized greats in computer science. 

From Lysenko Loonies to Climate Change Clowns

George Rebane

The 27jan12 WSJ publishes a major piece ‘No Need to Panic About Global Warming’ signed by sixteen science and technology heavyweights.  From my perspective as a systems scientist, I long ago joined the once lonely voices that pointed out that the climate change alarmists (especially of the anthropogenic global warming stripe) were intellectually naked shills with a pernicious political agenda.

Led by internationally recognized serious science commentators (links on right) like Anthony Watts, and locally by Russ Steele, the voices of reason and scientific accuracy have been growing and becoming more prominent over the last years.  But the political steamroller has yet to show significant slowing.  AGW is now a recognized and much applied tool of the political Left which uses it at every turn to remove individual liberties and shackle economies.

The authors of the cited piece point out the eerie similarities of how politicized science today recalls the days of church proscribed scientific enquiry and communist mandated politically correct sciences of (Lysenko, et.al.) of the USSR during the first half of the 20th century.  We are there again with claimed “incontrovertible evidence” that AGW exists.  Science contains no incontrovertible evidence, and once ‘evidence’ is so ascribed, it exits the halls of reasonable enquiry and science and enters the temples of religion.

The piece concludes –

A recent study of a wide variety of policy options by Yale economist William Nordhaus showed that nearly the highest benefit-to-cost ratio is achieved for a policy that allows 50 more years of economic growth unimpeded by greenhouse gas controls. This would be especially beneficial to the less-developed parts of the world that would like to share some of the same advantages of material well-being, health and life expectancy that the fully developed parts of the world enjoy now. Many other policy responses would have a negative return on investment. And it is likely that more CO2 and the modest warming that may come with it will be an overall benefit to the planet.

If elected officials feel compelled to "do something" about climate, we recommend supporting the excellent scientists who are increasing our understanding of climate with well-designed instruments on satellites, in the oceans and on land, and in the analysis of observational data. The better we understand climate, the better we can cope with its ever-changing nature, which has complicated human life throughout history. However, much of the huge private and government investment in climate is badly in need of critical review.

Every candidate should support rational measures to protect and improve our environment, but it makes no sense at all to back expensive programs that divert resources from real needs and are based on alarming but untenable claims of "incontrovertible" evidence.

The Value of Stereotyping

George Rebane

Stereotyping is the use of a template of characteristics that are thought to belong to members of a particular class more frequently than members of the general population of which the class is a subset.  The template of characteristics is also known as the stereotypical characteristics like, say, a plastic pocket protector full of writing instruments more often seen in the shirt pockets of male engineers than in the pockets of other men.

Stereotyping has gotten a bad rap in our society, and its use is roundly criticized in the public forum.  However, stereotyping, so as to assign or exclude someone from membership in a given class, is built in to almost every critter that lives on this planet.  Why?  Simply because it is a low cost survival technique that has evolved in all species to simplify quick decisions about the famous ‘Three F’ functions – feed, fight, or mount - important to everyone.

But if you understand the underpinnings of stereotyping, it can be a valuable tool in making all kinds of decisions, and also serve as a mirror for better understanding yourself.  I was motivated to write this piece when recently reading Daniel Kahneman’s monumental Thinking, Fast and Slow which dances around the subject without getting into the nitty-gritty of it because of the little math involved.  Kahneman, recent Nobelist and co-father of behavioral economics with the late Amos Tversky, is also a giant in the field of psychology.  In that field stereotyping comes up under the forbidding label of representativeness (q.v. – which is short for quod vide or ‘which see’, and I’ve concluded that its modern version is simply ‘google it’.)

Now consider seeing a good looking woman, finely coiffed, beautifully dressed, and dripping with expensive jewelry, getting out of a chauffeured limousine in front of a fancy restaurant.  You wonder if she’s a member of the currently notorious ‘1%’.  More particularly, what are the chances (friendly word for ‘probability’) that she is a member of that exclusive class, given she has some of the stereotypical characteristics of that class that you just observed?  If you had been facing the other way and your friend told you that a woman just got out a car behind you; then without seeing her, what would you answer if he mused whether the woman belonged to the '1%'?

Science still says, ‘Creation occurred.’

George Rebane

The best brains in cosmology ranging from Stephen Hawking of Cambridge to Alan Guth of MIT have been trying to munge the equations and the data to come up with some/any kind of support for the proposition that the universe did not have a beginning.

NewScientist.com reports in ‘Why physicists can’t avoid a creation event’ -

YOU could call them the worst birthday presents ever. At the meeting of minds convened last week to honour Stephen Hawking's 70th birthday - loftily titled "State of the Universe" - two bold proposals posed serious threats to our existing understanding of the cosmos.

One shows that a problematic object called a naked singularity is a lot more likely to exist than previously assumed (see "Naked black-hole hearts live in the fifth dimension"). The other suggests that the universe is not eternal, resurrecting the thorny question of how to kick-start the cosmos without the hand of a supernatural creator.

While many of us may be OK with the idea of the big bang simply starting everything, physicists, including Hawking, tend to shy away from cosmic genesis. "A point of creation would be a place where science broke down. One would have to appeal to religion and the hand of God," Hawking told the meeting, at the University of Cambridge, in a pre-recorded speech.

After looking under every theoretical rock available today, and then some, the bottom line is that “all these theories still demand a beginning.”

As a proponent of Intelligent Design (not to be confused with Creationism), I continue to celebrate such corroborative reports as satisfying Occam’s razor to the max.

'The Next Grand Minimum' Launches

George Rebane

Russ Steele, Nevada County blogger extraordinaire, closed down his long running NC Media Watch blog.  He has moved his blogging enterprise to a new blog - 'The Next Grand Minimum' - that was launched today.  In this research and publishing effort he will focus on real climate change backed by science that is yet to be tainted by politics.  Specifically, he will present ongoing scientific evidence - imported and generated - that will validate or falsify the advent of the next grand minimum (NGM) in earth's ever changing climate.

Grand minima in precipitation and temperature have occurred cyclically throughout the millenia, and the next one appears due in the lifetime of those of us here now.  Russ will collect, develop, and present evidence that is relative to the advent of this civilization impacting event.

I and most RR readers wish Russ the best in his new and important endeavor.  RR has installed a permanent link to the blog in the 'Our Links' section in the right panel.  We expect many points to the published work on that site.

Break a leg Russ!