Kwabena Boahen: Making a computer that works like the brain

TL;DR

The brain is able to process information using 100,000 times less energy than computers, thanks to its parallel and distributed network of neurons.

Transcript

I got my first computer when I was a teenager growing up in Accra, and it was a really cool device. You could play games with it. You could program it in BASIC. And I was fascinated. So I went into the library to figure out how did this thing work. I read about how the CPU is constantly shuffling data back and forth between the memory, the RAM and ... Read More

Key Insights

• 🧠 The brain is able to process information using 100,000 times less energy than computers do currently.
• 🌐 The brain processes information in a parallel and fluid manner, compared to computers which process information in a serial and rigid manner.
• 💻 As transistors continue to shrink in size, they will start to behave like ion channels in the brain, leading to mixed up ones and zeros and potential computer crashes.
• 🔌 Computers require a large amount of energy compared to the brain, with the fastest computer in the world consuming the same amount of electricity as 1,200 households.
• 🧠 The brain's ability to store and process data in a distributed and redundant manner makes it inherently robust, unlike computers which store data locally and are more brittle.
• 🌍 Kwabena Boahen believes that incorporating African perspectives and ideas into computer technology can lead to new and innovative solutions.
• 💡 Kwabena Boahen's team has designed a silicon chip that models the retina, which extracts different types of information just like the brain.
• 🎶 Kwabena Boahen envisions putting "Africa in the computer" to create a new kind of computer that is creative and imaginative.

Questions & Answers

Q: How does the brain's energy consumption compare to that of computers?

The brain is able to achieve efficient information processing with just 10 watts of energy, equivalent to a laptop, while computers like Blue Gene consume 1.5 megawatts of power.

Q: What is the key difference in information processing between the brain and computers?

Computers follow a serial and rigid process where data goes through a bottleneck in the central processing unit, while the brain's neurons and their connections create a parallel and fluid network for information flow.

Q: Why is the brain considered more robust compared to computers?

The brain's distributed data storage and redundant connections make it more robust, allowing it to recover missing information even when there are errors or disruptions in the network.

Q: How is energy efficiency in the brain related to the future of computing?

The research aims to build computers that mimic the brain's efficient processing methods, which could lead to more energy-efficient and innovative computing technology.

Q: What challenges do computer technology face in terms of device size and computation?

As transistors continue to shrink, they start to behave like neurons in terms of sporadic current flow, leading to traffic jams and errors in computing. The brain provides a potential solution for computing with these small devices.

Q: How did the researchers model the brain's efficient processing in a silicon chip?

By reverse engineering the retina, which is a piece of the brain that lines the inside of the eyeball, the researchers used the network of neurons and their connections as a blueprint for designing a silicon chip that mimics the brain's processing methods.

Q: What were the benefits and limitations of the artificial eye using the silicon retina chip?

The artificial eye provided sparse snapshots of information, similar to how video compression works, but the researchers were able to reconstruct the image. This shows the potential for developing computers that work more like the brain in processing visual information efficiently.

Summary & Key Takeaways

• Computers were once believed to be much faster than the human brain, but they consume a significant amount of power compared to the brain.

• The brain's efficiency lies in its parallel and distributed network of neurons, allowing for dynamic and adaptive information processing.

• The future of computing could involve building computers that work more like the brain, using the brain's efficient processing methods.