TCS Siruseri Campus

Today am visiting the TCS Siruseri campus in Chennai. Very elegant & interesting structure built by an Urguan Architect.

• The legend has it that while discussing the architecture, Carlos drew butterflies & that stuck as the theme.
• A majestic 5-floor-high open Atrium corridor forms the spine of the butterfly with 6 buildings forming the wings, the buildings themselves are butterflies w/ a small spine elevator bank in the middle and the two north & south wings buildings.
• The side view below shows three buildings EC1(right),EC2 & EC3. I am in EC3.
• The spine atrium on the far side has a pond, benches, shops & interesting restaurants – The Saravana Bhavan food is exotic. The spine even has a Subway !
• The campus hosts > 20,000 associates

The second picture below shows the full 6 buildings and the observation tower (which is still under construction)

Scaling Big Data – Impermium

Came acorss an informative blog on scaling big data – “Built to Scale: How does Impermium process data?” Quick notes from the blog:

1. Don’t fall in love with a technology so much that you cannot be separated – Be flexible in scaling as you grow

   • “Parting is such a sweet sorrow”, but change is an essential component of an infrastructure at scale 
   • The technology selection and consumption should be a continuous process, introducing new technologies as needed by the growth. I found Impermium’s path from grep to Solr to Elastic Search very illuminating; I have done the same before.  

2. Technology needs are not static

   • A corollary of #1 above – Growth on all parts of the stack will not be uniform.
   • For example Impermium found scaling challenges in search and they moved to Solr & then to Elastic Search

3. There are no perfect technologies

   • If you are doing interesting work, be ready to tango with open source code. This is essential – I also found this to be true.
   • Even if you don’t plan to change the code, many times deep understanding comes from reading the code

4. Select technologies that you can dance with

   • The flip side is that one should select technologies that you are comfortable working under the hood.
   • In my case, while I love Erlang, I am not that comfortable with that language. So given a chance, I will go with Java or Scala

5. Benchmark is nothing but a story in a specific context

   • So true. Benchmarks are transitory & personal.
   • Understand them, but they need not be true for your transforms, your data model and your processing.
   • Benchmark early & benchmark often … with your scenarions, models, transformations, mapreduces & data

Thanks Young for the short but very interesting blog. Keep up the good work …

Cheers

<k/>

Kareem Abdul-Jabbar: 20 things I wish I had known when I was 30

Kareem Abdul-Jabbar has an excellent blog at Esquire on 20 pieces of advise to the younger self at 30. Thanks to Jason Hiner‘s tweet.

The blog & the comments are a must read.

A few of them hit home for me:

• Be patient
• Listen More than Talk
• Being right is not always the right thing to be
• Do one thing every day that helps someone else.
• Do one thing every day that you look forward to doing. 
• Don’t be so quick to judge.
• Everything doesn’t have to be fixed.
• Play the Piano
• Become Financially Literate

Ref: Wallpaper from http://www.basketwallpapers.com/USA/Kareem-Abdul-Jabbar/

Data Science Engineers – The new breed of Data Scientists ?

While there is lots of interesting discussions about Data Scientists, or lack there of. The role of Data Science in Big Data is well understood. I think the need is actually for Data Science Engineers. I had a set of pictures explaining this concept and interestingly came across a blog by HortonWorks on the topic of Data Scientists.

[Update May 19, 2013] An informative blog in the Wall Street Journal about Data Scientists by IBM’s Irving Wladawsky-Berger – Data Science is a multidisciplinary evolution from business intelligence & analytics. In addition to having a solid foundation in statistics, math, data engineering and computer science, data scientists must also have domain expertise.

What says thee ?

Google Doodle on Euler’s 306th Birthday

 

Good work guys …

Is our Neocortex a Giant Semantic Bloom Filter ? Of Natural Intelligence, Machine Learning & Jeff Hawkins

L’Apéritif:

In a set of four lectures spanning about 3 years, Jeff Hawkins explains how & why big data can only be solved by evolutionary-adaptive-continuously-learning models incorporating principles from the working of Neocortex.
It does make sense – especially for NLP, NLU & Knowledge Representation. I am a big fan of the Borgs and their coordinated intelligence.

These are my annotated picture-notes …

L’Entrée:

Let me begin at the beginning. The other day I came across 4 very interesting talks by Jeff Hawkins on Biological Inspired Machine intelligence.

Call it serendipity because we have been looking for more effective ways for Knowledge Representation (KR) & Natural Language Understanding (NLU)

For example movie names, while very easy for humans to understand, a MaxEnt NER finds it very hard.  Knowledge Representation & Association is more harder !

We are experimenting with a few techniques like word-based tries (ie. spell-check sentences by words), higher order federated Bloom Filters and n-gram hashing. Planning to incorporate some of Jeff’s ideas …

I digress … Topics for another day … back to Jeff & Machine Intelligence …

Very inspiring, extremely thought provoking talks – as usual the inimitable Jeff Hawkins at his best

1. Google Tech Talk : Jeff Hawkins, “Building Brains to Understand the World’s Data“
2. UC Berkeley Graduate Lectures
   • “Intelligence & Brain : recent Advances in Understanding How the Brain works” by Jeff Hawkins
   • “Intelligence and Machines: Creating Intelligent Machines by Modeling The Brain” By Jeff Hawkins
3. “Advances in Modeling Neocortex and its impact on Machine Intelligence” by Jeff Hawkins,  Smith Group Lecture presented at the Beckman Institute for Advanced Science & Technology at the University of Illinois at Urbana-Champaign

Le Plat Principal:

The four talks have lot of depth and are packed. Moreover Jeff talks very fast – I listened to the talks a few times – at least 3 hrs per one hour talk. You should listen to them slowly & rewind as reqd. It takes a few hours to get one’s head around the various ideas.

Let me annotate a few of his slides – those I was able to internalize to some extent:

Focus & premise[3]:

The assertion, that many problems can only be solved by incorporating principles from the working on Neocortex, is interesting.

BTW, it does make sense – especially for NLU & Knowledge Representation.

As Jeff mentions later, the behavior need not be human-like, but the representation, interpretation & “understanding” would be.

Neocortex Architecture[3]:

“Neocortex is just a sheet of cells  2mm thick, the size of a dinner napkin” – Amazing what it can do!

The Six Principal Essentials of Biological Intelligence

The picture says it all.

Learning involves training and adaptive connections

The concept of streaming events & the learning mechanisms

Patterns from complex data streams

The paper “Hierarchical Temporal memory” has the gory details about the Hierarchical Temporal Learning.

Future

Interesting observation: Emotion, the fundamental aspect of being human, is not a requirement for intelligence – reminds us of Spock, of course.

Machine intelligence is not about replicating human behavior or even passing the turing test. I agree on this – we need the machines to think & do things we cannot do thus augmenting us. Make us stronger where we are weak !

Le Digestif

What interested me most was the sematic knowledge representation, NLP & NLU. The ability to understand and store concepts, the capacity to generalize as well as the mechanisms of strengthening and weakening connections based on external signals – just beautiful …

Agree that the Sparse Distributed Representation could be the language of all the intelligent machines.

The SDR looks a lot like a giant Bloom Filter

The planes can be considered as rows and a column as the temporal dimension of the semantic mapping (the memory of sequences). Which equates to a giant n-dimensional Bloom Filer – a data structure we can grok (Pun intended as Jeff’s product is called Grok!).

The bloom filter analogy, while extremely simplistic, is conceptually congruent, in the sense that “similar values have similar representation”, of course depending on the hash algorithm.

After listening to the talks and thinking them over, I have a thousand questions in many directions. I will post the answers as we develop this through for our needs. Please send in your insights as comments to this blog. AM sure it will help a few folks !

1. How do we handle semantic categories ? 
2. How do we build more sophisticated representations based on spatial patterns ?
3. What is the hash function that maps a slice of semantic to this giant Bloom Filter ?
4. How does it handle collision? Corruption ? Clustering for resiliency/self adjusting representation ?
   • Collision might be good and I think that is what Jeff calls as semantic generalization
5. How does the semantic slice mapping function differentiate between a search & computation to trigger appropriate actions?
   • For example the following two questions require different actions: 
     • “What is stock price of IBM ?” vs.
     • “What is the volatility as reflected in the beta of IBM for this quarter ?” 
     • The first one is a search while the second has computation …
6. Is the hash function same for all of us or is it different for each person ?
   • Most probably the function is a learned artifact.
7. Another interesting vector is the Hierarchy & higher patterns of temporal coalescence/slowness – the high-order capability, tweaking the learning rates across the layers.
   • How can this be modeled with the analytical data structures we have?
   • And what are the mechanics for stable representation of pattern sequences – because with dynamicity and temporality comes the difficulty of snapshots and consistency between them.
   • The unique representation of the same sequence, at a later time in context of the earlier invocation is interesting …
8. How do we “put a classifier on the top” ?
   • Play with permanence? Probability?
9. What are the algorithms to prevent run away prediction?
   • I agree that we could account for rapid state difference vs. slower state; we still will have to encapsulate it in some form of code

Finally, can we build “Amazingly Intelligent Machines™?” Yes We can !

And agree with Jeff that “It is essential, for the survival of the spices, that we build them” …

The Big Data Convergence

As we scan the concepts, technologies, products and the practices in the big data space, lot of things get muddier.

Neither the progression nor the boundaries are clear. We are still in the descriptive stage in terms of the application of the analytics technologies.

I had a good conversation with Bob Friday yesterday – his question was “What prevents us from answering 80% of the questions via automatic inferences ?” And that is the “Adaptive” stage we need to be …

I think a diagram is much better than me writing 100,000 words. So here it is :

In many ways, a lot of the underlying technologies are converging.

For example, A(rtificial) I(ntelligence) = NLP + N(atural) L(anguage) U(nderstanding) + ML + K(nowledge) R(epresentation) + Reasoning

Are Amazing Intelligent Machines in the works ?