This video session features the keynote speaker Professor Geoff Hinton FRS, “Deep Learning”. This lecture was filmed on May 22, 2015.

Watch full video at …

Background and Approach

This blog is based on my talk in London at the Re.work Connected City Summit on Deep Learning Applications for Smart cities. The talk is based on a forthcoming paper created with the help of my students atUPM/citysciences on the same theme. Please email me at  ajit.jaokar at futuretext.com  or…

If you follow news about the Internet of Things, you will have read quite a few articles that attempt to predict the number of connected devices by the year 2020.

http://www.m2mdaily.com/wp-content/uploads/2015/09/IoT-Device-forecast-5.png

The chart displayed above,…

Have you ever wondered how to segment your customers? Customer segmentation is a really useful technique to group similar customers together and understand what works for that. You can then tailor your offering and marketing messages to the specific segments. If you do it right, you should be able to see a healthy increase in sales. After all, companies like Amazon target their customers on an individual level so you should at least be targeting them on a segment level.…

Codified narrative is the product of converting human-friendly narrative into computer-friendly code. In past blogs, I discussed my own approach towards this process of codification. Here, I will be covering the idea of spatial, temporal, and contextual distribution of codified narrative. I have never suggested that narrative can or should be used in place of quantitative data. However, I have reflected on how the quantitative regime has tended to dominate discourse; this has perhaps led to…

Some of the rarely shared trade secrets in machine learning:   Original post: on linkedin

1. Bootstrap sampling & the magic number 0.63

Even though randomly sampled,…