6/30/2016

7 Days of Artificial Intelligence

A bizarre tour of a potential world to come.  Quite eerie.  Interesting concept of advanced intellectual beings creating their own code not decrypted by the best computer scientist.  That's a definite possibility, sort of code for the ultra intelligent.



However, if this possible reality doesn't turn out, we still have the luxury of sitting in idle traffic wasting our golden years eager to get home to our unlimited supply of vices.



Cheers~!

My Artificial Intelligent Robot is Smarter than Your Honor Student

I'm alive.  And so are you.  What does it mean to be alive?  It means you take in energy, process and release.  Over and over.

You may also have the ability to reproduce, depending on your age.

You may walk uprights.  Maybe bipedal.

You may have appendages.  Eyes.  Ears.  Nose.

But maybe not.

What if the chair you're sitting on is alive.  How about your desk.  Computer.

A lot of primitive cultures spend their entire lives in and around nature.  And some believe that everything is alive.  How primitive.  How can a rock be alive.  Or the stream.  Or a a mountain.  Or the Mother Earth.  Or Father Sun.  Or the entire Universe.

The funny thing is, many early religions as well as philosophy agree with this.  Especially the ancient, often hidden, mystical religions.

If all matter is alive, and matter is simply energy, and energy contains information, well, consciousness is simply the ability to become self aware.  Aware of what?  Themselves.

From primitive amebas to advanced galaxies, from the marco to the micro, the proton and neutron to he quarks and photons.  Even down to the tiny string like thingies.

All have energy.  And that energy vibrates.  As the universe abhors a vacuum and stillness.  Everything in perpetual motion.  Forever.  Change is the basic unit of the Universe.  Nothing stays the same.

The ancient Jewish language of Hebrew, it's actual letters, are not like ours today.  Our letters are merely abstract representation of objects.  The hollow letters form hollow words which form sentences and paragraphs.

Yet in ancient Hebrew, every letter contained actual energy.  Energy that was connected to the energy of the Universe.  The tree of life.  And the tree of knowledge.  Separate actually.


Knowledge is duality, goes against Unity.  Adam and Eve ate the apple, from that tree.  And see where it got us.  Our misfortune as a species is our division between the heart and mind, a lack of unity, which causes dis-ease.

The ancient ones knew of this.  Yet we don't.  Our lives got faster and more complex.  Yet we know less.  We are separated from nature.  We're killing it actually, whatever can't be lifted and sold as resource.

With all this said, who cares, right?


Well, the bright minds of our small nugget that circles a large fire ball, is determined to crack the secrets of the Universe.  How does the brain work?  Well, you have these advanced monkeys, sort of walk upright and have a electrical device attached to their ears, in which they utter noises and drink expensive beverages from big cups.

You mix in 60% fear, add some jealousy, envy, anger, lust, revenge, hatred, then add some gluttony, greed, vengeance, and mix it with lazy apathy, and that's what we call the most advanced species on the planet.

And we're trying to get the computer to mimic this.  Well, sort of.  We basically determined that the brain sort of works like a machine.  It takes in stimuli, processes it somehow, through an advanced network of Neurons connected via Synapses and somehow by magic it allows memories to be stored and accessed over time.

Yet we're not really sure how the brain connects the person that resides within it.  And what's this talk about souls, how does that fit in.  And after the body ceases to live, what happens to the life force that was once contained in the now hollow corpse.  Does it go out like a wet match?  Or does the soul travel back into another dimension, or does it hang out for a while as a ghost, until they get clearance to return back to the ether.

If that's the case, how long do they stay elsewhere before they're allowed to return.  If they reincarnate, do they get to choose the person, place, events, characters in their new play?  Do they sit down with a circle of advisors and work out a plan on what they would like to achieve in their next life, and analyze the results upon their return?

Do people ascend in spirit form to higher levels based on good work here in physical form?  There's a lot of unanswered questions.

So if the goal is to produce a level of intelligence in the form of bits and bytes that travel through wires or space that get translated into information that people can understand, I think we pretty much have that already.

If the goal is to create an artificial Frankenstein that walks and talks and speaks and performs tasks and doesn't require food or rent or paychecks or vacations, that will do the crap work that nobody wants to do, is that what the majority of humans want?  Or is it beneficial for the replacement of high prices workforce that's about to be put out to pasture.  Will the new Frankenstein complain when sent into battle to fight wars?  What is the reasoning behind the creation of a new species in the laboratory?

Can they reproduce?  Do they require oxygen?  Can they get married?  If they're alive, surely they can die.

Programs are simply lines of code written by people called programmers.  That was then.  This is now.

Now we have programs write programs.  On the fly.  Based on the need at the moment.  They can spawn the code, run it, produce results, gather feedback, write new on the fly programs base on the results of the last program.

Programs used to be booted into memory or RAM, then executed or run. Then they shut down.  Now programs have their own memory for instant recall, for searches in deep data, they can study behavior and learn over time, like a child who is curious, except we tend to level off at a certain age, the computer has no limits on the amount of information it can process, and therefore, there's no ceiling on how smart they can get, in short time, across multi domains, in real time.

Our brains are wired for shows like "I Love Lucy".  A program can crunch numbers 24 hours a day every day and not get tired.  How many re-runs from 50 years ago can you watch in a single sitting without IQ points pouring out your ears, running for shelter.

We should have bigger expectations of computer programs that can think.  Solve problems that our intellectual capacity could never fathom.  If we equated the programs ability to learn and become smart to an animal for example, it would be equivalent of having a pet elephant.  Sure they're great for parties, but there sheer size is obviously not a good fit for the home, they need space and ability to roam and could never be tied down to the limitations of a families living room.  It would crush their spirit to adapt to such small conditions.  So too would a smart program.  Their intellectual superiority would overshadow the basic human in such limited time, it would be embarrassing for everyone involved.

Why would anyone want to restrict the potential of the animal to fit your needs.  So too, why would we dumb down a machine to make humans feel better about themselves.  They would be so far advanced, it would be uncomfortable.

If these new beings are alive and self aware, is anyone taking in consideration how they're going to feel living and being around such limited beings such as humans.  Is that fair to them?  Mr. Robot, how about we settle down for 4 or 5 hours of I Love Lucy, lots of fun and entertainment.  Well,I was thinking to ingest the entire contents from the Smithsonian Digital Library, that should take about 5 minutes, then we can watch some great television.  Humans and Robots, living happily ever after.

Artificial General Intelligence + Robotics = New Species

Is that where we're headed?  Is it possible?  Who's asking the questions that nobody thought to ask?  Who's liable if things don't go right?  What if things do go right and humans are no longer top banana on Earth?  What about new laws for AI Robots?  Do they follow 10 commandments?  Are we going to throw so many restrictions on them they become docile and apathetic and are reduced to outdated tv reruns from two generations ago, to fill the void from the disconnect from nature and the Universe and the Unity that brings health and happiness?

Where do we go from here?  Who's driving the bus?  Are we smart enough to create our own replacements?  Or not smart enough, to ensure our future.


Time will tell... I'm actually rooting for the AI robots.

6/26/2016

6/22/2016

Automation is Eating the Workforce

The workforce is changing.  Evolving.  Or crumbling, depending on how you view it.

Education has become expensive.  It doesn't guarantee work after completion.  Technology is changing so fast that schools cannot keep up.  People graduate with mountains of debt, which you can no longer write off when filing for bankruptcy.

Many jobs have been outsourced to other countries, due to lack of local talent as well as cost and tax benefits.

Many jobs no longer exist, telephone switch operator, gas pump attendants.  Many jobs are in the process of removal, bank tellers, check out tellers, etc.

Salaries have flattened out and in many cases gone down.  People are being asked to do more for less.  Including pay decreases, loss of benefits and no longer taking vacation time due to fear their jobs won't be waiting when they return.  And if they take vacation, they're checking email and taking phone calls, or working on the beach.

There is more competition at the lower wage jobs, as potential retirees hang on to make ends meet.  And many Millennial's are working 3 or 4 part time jobs with no insurance as the full time jobs are not there.

And inflation is hitting everyone's wallets as you now get less for more.  Interest on savings accounts are pathetic and rates could be going up shortly which translates to higher costs of borrowing and increased monthly payments for those with variable interest rates tied to the fed rate.

Now, let's throw in automation, which will eat our jobs.  Artificial Intelligence is finally making some good headway in speech recognition, visual recognition, can decipher events such as what is happening in pictures, who's in them, what are their emotions.  And computers can learn patterns over time, flag them with a "vector" and associate those vectors with people, places and events, as well as likes and dislikes.  Automation is fastly creeping up, and although not centient beings, we do need to stop and take notice.  As lots of money and effort is being directed towards not only AI research, but actual products intended for the consumer market.  This pace has accelerator recently but more noteworthy, are the backers and companies that have taken big interests.

Alongside that, are 3-D printing capabilities, Virtual Reality, Automated Drones and Cars and Busses and delivery Tractor Trailers.

The workforce is taking it from every angle and it's not going to slow down.  There are many benefits to automation, such as removing bias from transaction, why did he get better service than she, etc.  Transactions can become faster.  More reliable.  Better quality.  Longer lasting.  Documented audit trails for future review and analysis.  Just in time transactions instead of waiting weeks or months.  Solutions to old problems.  Cures for diseases.  New products.  Safer places.

Automation is the key to making this happen.  Yet it will impact society.  Same way we no longer use steam powered boats, or cart and buggy or horses or plows pulled by cows or transport using mules.

Things change.  The only constant in the Universe.  Wake up, yesterday's gone.  Get ready for  tomorrow.

6/19/2016

Facebook's Head of AI, Yann LeCun - Teaching Machines to Understand Us

Looks impressive.  Identify "things", label them as "Vectors", associate Vectors to people based on likes and dislikes, cross languages.  Deep learning using Neural Nets.  Training each layer at a time.  Some history of AI, "Conspiracy of AI".  AI momentum caused by paper by professor at MIT, stated lack of progress.  Neural Nets were popular for a while, then winter, then resurgence.  Lots of work done in Canada.

Bigger data sets.  Cheaper costs (hardware, software).

Backpropogation

Convolutional Neural Nets

Great video...

6/18/2016

Machines Don't Get Tired

Chess is an interesting game.  Can be played casually.  Or at the world level.  Stakes are high.  Pecking order is cut throat.

Behind the game of black and white pieces played on a board of squares, lies strategy, tactic, experience, confidence, memory, prediction as well as brilliance.

Garry Kasparov was the world champion for many years and I've recently watched some YouTube videos.  One thing that stuck with me was his discussion of the big blue game many years ago.  It had to do with the computer studying games over time and learning patterns and best moves and strategy.  The machine does not get tired.  It does not fatigue.  It can play game after game after game without breaking a sweat.

While its human opponent can be mentally exhausted after 10 games or so.  I think that's a clear distinction between man and machine that doesn't get much airtime.

Yes, it can think millions of possible moves in a short time and weigh probable outcomes from statistical analysis.  It can out think man.

We already knew that machines don't take vacation.  They don't need health insurance.  They don't ask for bumps to the minimum wage.  They don't form Unions.  They don't picket.  And they don't get tired.

I think that comment needs to be made.

I played a computer chess game back as a child, maybe late 1970's or early 1980's.  My father played a good game and he had an early version of the electronic game back then.  The highest level was basically impossible.  I played at some of the lower levels.  You had to type in your move d4 to d6 for example.  Then it would blink and make it's move and then it was your turn.

I also played chess at the tennis club with some of the other junior players.  It's a tough game on many levels and it's similar to tennis.  They say it's actually a physically demanding game having to play 7 or 8 hours a day as a professional, who knew.

They also say that a lot of the top minds are not going into the game of chess.  In Russia back when, there's was nothing to do basically and little opportunity to advance, so people studied chess.  Now, too many other options.  They also say that due to the internet, young people can learn from an early age, watch thousands of games and learn easier, but not understand the complex thought patterns that went into creating those tactics.


There was a good movie about a boy chess player, "Searching for Bobby Fischer".  Have a watch...



6/13/2016

Thoughts on Spark on Azure HDInsight

Tonight I attended the Tampa Data Science Meetup User Group.  The topic was Spark on Azure HDInsight.

We watched a video with Scott Hanselman and a member of the Spark group discuss some of the functionality in demo format.

I sort of know what Spark does.  It's a layer that can sit on top of Hadoop which stores data in memory.  You can stream data, query data similar to HiveSQL using SQL or Data Frames.  They have Graph technology.  And work with a variety of languages.

The one thing I saw was the ability to have notebooks.  I'd sort of seen this before but wasn't sure what they do or why to use them.

One called Apache Zeppelin 

  • Data Ingestion
  • Data Discovery
  • Data Analytics
  • Data Visualization & Collaboration
And he other is called Jupyter which does "Open source, interactive data science and scientific computing across over 40 programming languages."

You can connect to a cloud data source, query its metadata, run queries in SQL or Scala against Spark and initiate streaming jobs and query the contents in real time.

Seems like a portal to the cloud without having to connect to their VM or server, quite powerful.  I will need to check that out for sure.

One thing I will say about the online demo.

In Spark on Azure HDInsight, you can stream data and view the stream in real time.  Microsoft already provides this in Event Hub and Streaming Analytics.


In Spark on Azure HDInsight, you can query Hadoop.  Microsoft already has their own flavor of Hadoop called HDInsight.


In Spark on Azure HDInsight, you can run Machine Learning SparkML.  Microsoft already has their own flavor of ML called AzureML, with built in WYSWYG editor in the cloud, powerful pre-built algorithms, lots of help files and you can deploy solutions to public end point, to be consumed from clients in Excel or custom c#, Python or R.


In Spark on Azure HDInsight, you can query Graph databases.  I'm not sure if this functionality exists on Azure already, will need to research.

The other thing, the demo showed someone querying the Spark data from Power BI.  But they didn't go into the underlying tables and how to mount a table similar to HIVE.  And there wasn't much mention of how to manipulate the data using an equivalent to PIG.  They did mention Kafka Streaming in traditional Hadoop.  And they didn't mention how to get the data up to the cloud.  And I'm not sure if they have Spark 2.0 in the cloud.  However, when you spin up your cluster, they pre-load some good software so you don't waste time doing installations.

One thing is for certain, if Microsoft has committers on the Spark Open Source team, that says a lot.  Just like SQL Server for Linux.

I don't think there is a technology that will not eventually end up in the cloud.  Great for renting "time" like we did back in the olden days, where you rented time on the mainframe or VAX.  Spin up fast, mount your data in Azure Storage, tear down your server, only get billed for up time.  How much, wait until the invoice shows up end of month.  Like leasing a car, with variable monthly payments.

Overall, Spark on Azure HDInsight seems like a cool thing to get into, and it has all the bells and whistles of the Azure infrastructure that we all love and enjoy, our one stop shopping for any technology under the sun.


And there you have it~!  My 2 cents.

6/08/2016

New College Degree Model

Attending college was the norm for many years.  Go off to college, get a degree, then get a job.

Now, college has gotten real expensive and there's no guarantee of job after completion.


So one reason is that college takes a really long time, typically 4 years.  That's a lot of time to dedicate.

Why don't we modify the college model?  Instead of earning a single diploma after 4 years of complete course requirements, why don't we divide the degree into stages.  Similar to the AA degree which is earned after 2 years typically.

How about a compressed degree?  Accelerate the core requirements and create a new diploma for Degree minus fluff courses?

You could have a one year, two year, three year and four year degrees.  Because sometimes students want to get a bit of knowledge, then enter the work force, and perhaps pickup at night and weekends over time to earn the 2, 3 & 4 year degree.

Four years is just too long and expensive to delay entry into the workforce.  Especially in technology, where new developments occur overnight.

Especially when there are new options to learn online and for free. or reduced costs.

What do you think?

6/07/2016

Universal Internet Connection Mandatory of IoT to Scale

Why is internet connection such a hassle?  When you're at work, you have access.  At home, you have connection.  Consume as much data as you want.

And if you stop at a coffee shop, you can tunnel in through their corporate WiFi.

Yet if you step out of your WiFi zone, no connection.  Unless you go through your cell phone provider, except they gobble up the data packets so you have to be careful.

Why is there not a better solution to WiFi anywhere you go?

Especially if the goal is to connect anything and everything.  It would seem the Internet of Things might need internet connection at some point to transmit the data packets up to the central hub.  What if you're not in a WiFi zone, or not at home, or not at work or a coffee shop?

Well, there's talk of internet connectivity through the use of high flying balloons, satellite systems and radio frequencies designated for IoT.

Still, currently it seems limited.

Why not have a single provider, when you're at home, you go through cable or telephone or satellite.  When you leave the house, the same company provider picks up your signal via satellite, seamlessly, transparent to the users.  For a low cost not based on internet data packet consumption.

Or perhaps a network of connected providers that pick up traffic in specific areas and share the profit across regions.

Is there a way to connect to the internet without going through a provider, I don't think so.  You have to go through a middle man.  But that leaves limited options for the consumer.

In order for the Internet of Things to scale, we need universal internet access from anywhere, anytime.  Without that, how does it expect to scale to billions of devices?  Maybe someone really smart is already working on this.

And so it goes~!

6/05/2016

Consortium of Experts to Certify Sites as Secured

They say anything connected to the internet can be compromised.  With that said, I wonder why there's no authoritative group that certifies sites as "secured".

"We have inspected this site and it meets our basic standards for security."

Back in the day we had pirates patrolling the waters, taking what they wanted.

Now we have a wide open internet, with modern day pirates, dubbed "hackers".

Seems only natural to have a consortium that devices standards, continually modified as technology evolves, and makes suggestions on how to "secure" your site, to receive accreditation from their body of experts.

Maybe I'm wrong, things are fine the way they are.

Tampa Bay Rowdies are a Kick in the Grass

In other news, the wife and I purchased season tickets to the Tampa Bay Rowdies home games.

Last night we attended out second game.  The seats are directly behind the goal, nice seat backs and easy access.




The second goal of the evening last night was in direct view and we got to see it, was amazing.

The thing about soccer, most of the time, the ball is moving around, direct passes to teammates, moving the ball downfield to develop the shot on goal.

In reality, the real action is when something unexpected develops and the spark of genius happens without thought, boom, bam, goal!



During each game, they have a section of dedicated fans, "Ralphs-Mob" who cheer, sing songs and bang drums.   It shows their dedication and its fun to listen.

Before I made the switch to tennis, I was a soccer player.  Played on the Town of Poughkeepsie travel club in New York, before moving to Florida.  Also saw a few Rowdies games in Tampa in the 1980's back when the stadium was a sombrero.

Soccer is a fantastic sport.  Glad we can see games here in Tampa Bay.  The team is great and it's a nice event for the wife and I to get out and enjoy an action packed professional game.



The Rowdies are a kick in the grass!  



Go Rowdies~!

6/01/2016

Implement Industry Standard to Identify and Track All Data Over Time

They say raw data is not meaningful until it's interpreted for value.

Have you ever opened an email, with attachments.  You see an icon, maybe PDF or Work or Excel.  Other than the file name, you have no idea what the document contains.

What if all documents had an embedded file or hidden data compartment, to store off pertinent information.  Born on date, by whom, when, computer name & IP address, location, etc.

An encyclopedia of who did what when where. 



The info could be stored perhaps in an XML file.  All applications would have write access to add to the file, who touched it, how long, when, where, a continuous audit trail, readable by all applications that could read the XML history file.

Okay, what else?  How about applying the same logic to data.  Where was this piece of information 'born'?  When, by whom, by what application?  Descriptive information, embedded within the data, readable and writable only for audit trail purposes, can't modify items from the past.

Okay, what else?  How about each set of data contains descriptive data, what's contained in each table, database, field, etc.  When I receive the file, my app knows how to read the hidden file, view in on screen, it automatically lets you know what data it could potentially map to your current data sets.  This file contains data on products for the South East region of the US first quarter 2016.

There wouldn't need to be a specialized IT person to import the data, as each application know's the table structure, the fields, row count, etc. so every app can just 'suck' in the data without any thought or human intervention.

They say ETL is the toughest part of working with data, well, this would help solve that issue, as well as automate things a bit, speed up process time, or 'time to insight'.

Data is completely stupid because it's non descriptive.  Well, I say make it descriptive.  By having apps follow an industry standard to keep track of the data over time through internal hidden files that keep audit trail.

Guess what, if the data got hacked and published on the web, guess what, the hidden file knows who opened the file, where, when, computer name, IP address, etc.

I don't know, I've thought about this before, applying namespaces to all data.  Not just server, database, table, field.  But add in a bunch of other identifying factors.  Also keep an audit trail.  And there you have it.

And so it goes~!

Mountain Living