Arun Waves

October 8, 2016

Indus Valley Civilization – The Script (2)- literature survey

In the last few weeks I have surveyed numerous articles related to Indus Valley Civilization and its undeciphered script. These are great for gaining knowledge about the topic but not good for some serious scientific analysis using my latest tool set …………….. Machine Learning.

Here is a compendium (pics are from the related links, so all credits go to the authors of those links/articles);

1] Awesome pics:


2] Archeological Survey of India’s original publications: Memoir #77 in

3] Read up on the history, not just pics of script:

4] More pics:


5] Ancient writing systems:

6] Book with a catalogue of pics:


For some real papers:




Here are some concise relevant information;

  1. There are approximately 400 unique symbols
  2. Average length of an inscription is 5 to 6 symbols
  3. Maximum length of inscription is 26 symbols
  4. There is strong evidence that the direction of writing  was from right to left
  5. Most of the symbols are unique and do not repeat
  6. There is no cultural, political, religious, or linguistic ties to anything today, so no help from there
  7. No bilingual text has been discovered, like the Rosetta stone which helped crack the case of Egyptian hieroglyphs

August 7, 2016

Indus Valley Civilization – The Script

One of the few things that I impressed me in my high school History class was the chapter that dealt with the Indus Valley Civilization, a 4000 years old network of cities located in what is today the general area of India-Pakistan border. What impressed me was not how old it was but the evidence of advanced drainage and sanitary systems, broad roads that were straight and ran North-South/East-West, and the buildings that were deliberately laid out in a systematic pattern. I did not know the technical term at that time but here was clear evidence of Urban planning in a city that was built 4000 years old!!!


Indus Valley Civilization Script

The next thing that caught my attention were the numerous seals and tablets that contained interesting symbols that were sometimes accompanied by human and animal motifs. And guess what, this language has not be deciphered yet!!! I can’t help dramatize the whole thing ……….. imagine a person, 4000 years ago, picks up a hard and sharp tool, carefully inscribes these symbols which meant something to him just like these words mean something to you, then it gets used for many years and one day, someone lays it down one last time and then no one touched it for 1000s of years. What blows my mind is that 4000 years later, people anywhere on this planet can see this inscription from the comfort of their homes using technology that would be like magic to the person who inscribed this seal. And yet none of the 7 billion people can understand what was inscribed on that seal 4000 years ago.


What does it say?

Anyhow, snapping out of the dramatization, why did I awaken this dormant knowledge and mystery? Few months ago, during my routine nocturnal excursions or more accurately meanderings in the web, I came across an article in Nature ( which reported on a rare recent work done by Bryan Wells who is an archaeologist, epigrapher and geographer. He holds a PhD in anthropology from Harvard University and works at a university in Germany. I call it a ‘rare recent’ work since most of the work on the Indus Valley script is decades old and I believe the trail has gone cold. Around the same time, I was taking a serious interest in the field of Machine Learning and Artificial Intelligence especially the Neural Network part which leads to Deep Learning. Well, the nocturnal brain in its calm and meditative state, made a link between the two and I got hooked.

July 27, 2016

Neural Networks – scary good :-)

For the last year or so, there has been a steady increase in chatter about Artificial Intelligence/Machine Learning/Neural Network when it comes to anything online. After a brief stint with Fuzzy Logic during my undergrad, I never got an opportunity to dig into this topic further, other than keeping myself up to date with current events or milestones like the advent of IBM Watson (Jeopardy), Siri/language processing, self driving cars/various DARPA challenges, DeepBlue (chess), DeepMind (AlphaGo) etc. Almost by serendipity, I stumbled onto a Coursera course, Machine Learning by Andrew Ng (of Standford/DeepMind/Baidu fame) and found that my maths skills from my PhD in Quantum Mechanics fits right into this magical world of Neural Networks and Machine Learning. Cool, I dove right into the ocean of Deep Learning, layers,  gradients, features etc 🙂

Here is a ‘small’ result from the coding exercises of Andrew Ng’s Machine Learning course in Coursera. The task is to code a system that can recognize hand written single digit numbers and I am elated to report that Neural Networks is ‘scary good’ at it. What does ‘scary good’ mean? I did not have to explicitly code how numbers look, meaning I need not know Arabic numerals at all to develop a system that can recognize it from a non-standard medium (hand scribbled). Check out the results in GIF format below (Note: Yann LeCun’s LeNet5 has done this and more, way back in 1998);

Neural Network

Create a free website or blog at