The Brain Decoded: New Algorithm Bridges Mind and Machine

Where mind meets matter.

Computational neuroscientist Brokoslaw Laschowski and his team at the University of Toronto have developed an ML algorithm that teaches computers to interpret the human brain more efficiently,  allowing for better control of robotics and drones.

Brain-machine interfaces rely on decoding patterns of neural activity to translate thoughts into actions. Traditional brain-decoding algorithms struggle to generalize because brain activity differs greatly across individuals. The team's method avoids the common pitfall of “negative transfer,” where adding more training data can actually worsen performance.

“These findings challenge the dominant practice in machine learning, which focuses on developing and using large-scale datasets for training... Our study shows that quality, not just quantity, is important when selecting source subjects to train a machine learning model for brain decoding.”

— Dr. Brokoslaw Laschowski

Laschowski’s lab is now collaborating with aerospace engineer Hugh Liu to apply the model to mind-controlled drones and other adaptive systems that learn directly from brain signals.

Read more →