Month: July 2018

As you might well imagine, our family is a mini Big Bang Theory:  I am a chemical engineer married to a physicist with a son who is finishing his degree in computer science. Conversations around the dinner table can be almost anything from generational differences on political perspectives, to comparisons of communication styles, and of course the latest update on the video gaming world.

Last night we talked a bit about the media, which of course loves to cover things that will attract followers.  The editor’s rule, “If it bleeds, it leads,” sums up their fascination with violence and fear-based stories, and the song Dirty Laundry sums it up so well.  As we were finishing our conversation, I commented that energy is no longer on the minds of Americans, and I gave him a bit of history lesson about the legislation after the two Arab oil embargoes in the 1970s where gasoline was rationed and speed limits lowered to 55-mph.

He was shocked to hear that we repealed the 55-mph speed limit even though it clearly saves energy, money and lives because car wrecks are less fatal.  I gave him a short history lesson about the Fuel Use Act that forbade natural gas as a baseload fuel in power plants … and of course a bit about nuclear, which still powers our navy’s large ships.

As I was finishing this I commented that few today knew very much about how we got here, and like the Holocaust and other tragic events in the past, forgetting history almost certainly dooms us to repeating it.

I commented that today’s preoccupation with solar and wind seems to forget about the “rest of the story” to keeping the lights on and that the idea that batteries would save the day is a bit myopic and costly.

Finally, I suggested an informed dialogue on this to truly develop a national energy strategy and he blurted out: “Nobody cares about things like this.  People want to talk about health care, jobs, and sports.”

He may be right, but if that is true, we are in a very bad place and doomed to see history repeat itself.

Ali Rahimi, a researcher in artificial intelligence (AI) at Google in San Francisco, California, took a swipe at his field last December—and received a 40-second ovation for it. Speaking at an AI conference, Rahimi charged that machine learning algorithms, in which computers learn through trial and error, have become a form of “alchemy.” Researchers, he said, do not know why some algorithms work and others don’t, nor do they have rigorous criteria for choosing one AI architecture over another. Now, in a paper presented on 30 April at the International Conference on Learning Representations in Vancouver, Canada, Rahimi and his collaborators document examples of what they see as the alchemy problem and offer prescriptions for bolstering AI’s rigor.

“There’s an anguish in the field,” Rahimi says. “Many of us feel like we’re operating on an alien technology.”

The issue is distinct from AI’s reproducibility problem, in which researchers can’t replicate each other’s results because of inconsistent experimental and publication practices. It also differs from the “black box” or “interpretability” problem in machine learning: the difficulty of explaining how a particular AI has come to its conclusions. As Rahimi puts it, “I’m trying to draw a distinction between a machine learning system that’s a black box and an entire field that’s become a black box.”

I have blogged about this in the past. The idea that you throw all your data into a big box and process it to come up with true insights is largely hogwash. I have spent most of my 50 year career on this subject and proven over and over again that you must first start with a hypothesis and test it to see if there is a relationship. In most cases, as you do this, you find even what you thought was a simple relationship is more complex. Then, once you truly know what is true, you can begin to build a learning algorithm that will separate the data into groups of true and false.

If your organization has a big data initiative, please reach out to me and let me help you before your black box thinking kills your credibility.

We have had a running battle with these little critters, attempting to buy bird  feeders that claim to be squirrel proof.  I thought I was oh so clever hanging the feeder from wire fishing leader that was so fine I never thought a squirrel could climb down to the feeder.  And, for a while I was patting myself on the back … until today.

The picture tells the story.  The feeder supposedly has a wire mesh frame around it that allows birds to get into the feeder but was supposed to keep squirrels out.

Well, nobody told the squirrel.

Which reminds me of the phrase: “where there is a will there is a way.”  Perhaps it is time to recognize that this is the essence of the problem today with innovation and progress.  It is not about monetization or journey mapping alone … it is about the will to do something even when it is clear there is a compelling business case.

Finally, even the energy pundits are proclaiming the benefits of electricity over natural gas.  Yes, it does require technologies like heat pumps, infrared, and microwave to be employed, but the DOE’s stand against electricity as bad has finally been challenged.  In case you are unaware the DOE has been openly hostile to the use of electricity ever since its establishment out of the national wrath over the oil embargoes of 1973 and 1978.

I must point out that energy policies come and go.  Remember when we lowered the speed limit on highways to 55?  As I rocket along at 70 mph, I sometimes think about how wasteful that is.  But, of course I would not dare drive at 55 on today’s highways.  I would be run over!

But, our friends at Rocky Mountain Institute have formally concluded something I wrote about 20 years ago.  Nice to have my observation confirmed:

Read Energy Central’s article: Electrifying US Heating Needs Creates Massive Carbon Reductions, New RMI Report Finds

Of course, this is a bit too late to reverse the damage done, especially in the regulatory mood across most of the United States.  Yes, DOE and even RMI back then convinced the vast majority of our regulatory bodies that electrification was wrong and bad.