While Jeff Hawkins was creating the Palm Pilot, launching the era of handheld computing and amassing hundreds of millions of dollars, a big part of his mind was somewhere else: His true passion is something entirely different. It is an Einstein-worthy puzzle that has fascinated scientists for centuries: What is the source of intelligence?
Understanding the human brain and designing truly intelligent machines has been a lifelong passion for him. In his book On Intelligence, Jeff Hawkins takes a detailed look at how the human brain works, compares this to how AI currently works and explains his memory-prediction framework theory.
Numenta, his new startup is dedicated to create a new type of computing technology based on his theory described in the book. The product, called Grok aims to help us make sense of fast-flowing machine-to-machine data by recognizing patterns and building models. It is a cloud-based service that takes steady feeds of data from things like thermostats, Web clicks, or machinery. From initially observing the data flow, it begins making guesses about what will happen next. The more data, the more accurate the predictions become:
Grok automatically builds models of the data stream, then produces two types of output:
1. Predictions of future values, such as:
- How much energy will this building consume in the next hour or four hours from now?
- How much of this product will I sell today?
- Where will a web site visitor click next?
2. Anomaly and rare event detection, such as:
- Is this sequence of transactions unexpected?
- Is this machine in an unusual state?
Grok is still in limited release, with just a few customers in the fields of energy, media, and video processing. So far, the company claims, Grok has delivered results that are 10 percent to 20 percent better than various benchmarks, like revenue, optimal purchasing mixes, and machine servicing. The company expects to start selling Grok more broadly in the first half of 2013.