Just as the intricate symmetries in a kaleidoscope create mesmerizing patterns, so do symmetries play a crucial role in unlocking the secrets of artificial and biological general intelligence. Much like how physicists have discovered that certain transformations affect some aspects of a system while leaving others untouched, researchers argue that these same symmetries can guide us in finding the optimal representations for intelligence. This notion has already revolutionized physics, leading to a deeper understanding of the universe and even predicting the existence of new particles. Now, machine learning is embracing the power of symmetries, resulting in more efficient and adaptable algorithms that can mimic the complexities of biological intelligence. Furthermore, exciting advances in neuroscience demonstrate how the brain utilizes symmetry transformations for representation learning. The convergence of evidence across these disciplines suggests that symmetries may serve as a fundamental framework that shapes both biological and artificial intelligence, influencing everything from the structure of the universe to the nature of natural tasks. To uncover more about this captivating subject, delve into the full article!

Biological intelligence is remarkable in its ability to produce complex behavior in many diverse situations through data efficient, generalizable, and transferable skill acquisition. It is believed that learning “good” sensory representations is important for enabling this, however there is little agreement as to what a good representation should look like. In this review article we are going to argue that symmetry transformations are a fundamental principle that can guide our search for what makes a good representation. The idea that there exist transformations (symmetries) that affect some aspects of the system but not others, and their relationship to conserved quantities has become central in modern physics, resulting in a more unified theoretical framework and even ability to predict the existence of new particles. Recently, symmetries have started to gain prominence in machine learning too, resulting in more data efficient and generalizable algorithms that can mimic some of the complex behaviors produced by biological intelligence. Finally, first demonstrations of the importance of symmetry transformations for representation learning in the brain are starting to arise in neuroscience. Taken together, the overwhelming positive effect that symmetries bring to these disciplines suggest that they may be an important general framework that determines the structure of the universe, constrains the nature of natural tasks and consequently shapes both biological and artificial intelligence.

Read Full Article (External Site)