Brains And Computers At The VLSI Design Conference

One of the industry’s biggest events, the VLSI Design Conference, took place in Bangalore last week. This conference does a round-robin of cities, and this was the 10th time in its 33-year history that Bangalore was hosting it (the last time was in 2015). This year’s conference attracted over 1,800 technologist and leaders over five days – a huge turnout for this growing industry.

Incidentally, I have had the honor of serving on the Conference’s Organizing Committees three times: in 2011, 2015 and 2020. Being a founding member of the VSI Society of India, the parent body behind the conference, Cadence has been a sponsor for many years.

This year’s conference had many highlights, including having the Deputy Chief Minister of Karnataka, Dr. Ashwath Narayan, at the inauguration, and Member of Parliament for Bengaluru South, Tejasvi Surya, as the banquet guest speaker.

The legendary Dr. Rabaey
But probably the biggest highlight from the attendees’ perspective was the Banquet Keynote by the legendary Dr. Jan Rabaey. It was indeed a coup to secure Dr. Rabaey as one of the keynotes – not only had every electronics engineer in the room (and that constituted 95% of attendees) studied his text books, but this was a rare chance for attendees to hear him speak in person (he mentioned that he was last in India 22 years ago).

Dr. Rabaey is Professor Emeritus at UC Berkeley. He is a founding director of the Berkeley Wireless Research Center (BWRC) and the Berkeley Ubiquitous Swarm Lab, and has been the Electrical Engineering Division Chair at Berkeley twice. He has many areas of research interest, one of them being exploring the interaction between the cyber and the biological world. Indeed, the title of his talk at the Banquet dinner was “Of Brains and Computers.”

When Dr. Rabaey was introduced before his talk he was accorded a welcome that is usually reserved for movie stars! In fact, he said that earlier in the day he was inundated with requests for selfies, a testament to his celebrity status.

Difference between brains and computers
In his talk, Dr. Rabaey illustrated the similarities and differences between the human brain, or “biological computer” as he put it, and an electronic computer in a way that most of us probably have never thought about before. In this blog I’m going to summarize the differences that he presented.

The first difference is size. While the size of the brain has grown significantly over the ages, the computer has shrunk.

Second, the brain and physical computers have different goals. Computers were built to solve problems and to execute algorithms by performing a series of sequential steps. We built them to maximize performance while minimizing energy and cost. On the other hand, the brain uses a series of closed-loop feedback. We observe the outside world through the senses. We observe patterns that we put together to understand more complex patterns, and then we compare the things we have learned, and based on that we make decisions and perform actions.

Third, approaches to solving problems are different. Computers use an instruction-based paradigm – using an instruction set to sequentially execute a set of instructions to follow a task. The biological world, on the other hand, is based on a learning paradigm – observing of patterns and how they interact.

Fourth, the brain operates in a spatial-temporal space (i.e., data is collected across both space and time). “Spatial temporal” describes a phenomenon in a certain location and time.  In the case of the brain, certain areas of the brain only get fired up when the right patterns get presented. Most of the time they are in rest mode. On the other hand, computers are based on a regimented timing engine, where clocks determine when and in what sequence task/s have to happen.

In the rest of the talk, Dr. Rabaey went on to discuss the commonalities between the brain and computers (for example, both are at their limits, use of energy to perform functions, the effect of heat on ability to perform functions efficiently), the opportunities of converging the two, and neural design principles.

I was fortunate to sit in on a media interview with Dr. Rabaey during which he spoke about some fascinating research that has been and is being conducted at BWRC. I am sure there are a legion of engineers who are looking forward to seeing that interview in print soon.

Here’s a one-minute video of some glimpses of the conference.