Consciousness Can Be Mathematically Formalized, Says Leading AI Scientist
"We can describe any function using state transitions and all these state transitions using computational operators. If you agree that consciousness is a behavior of some kind, then it follows that we should be able to describe it in such languages," asserts Joscha Bach, challenging conventional notions that subjective experience resists mathematical description.
The AI researcher's ambitious framework for mathematizing consciousness represents a significant departure from traditional philosophical approaches, writes End of Miles.
Beyond Zeros and Ones
For Bach, current computational scientist at Liquid AI, the mathematical formalization of consciousness isn't about reducing experience to binary code but finding appropriate representational frameworks for describing complex mental phenomena.
"It's not necessarily about zeros and ones. It's about finding a representation in which you can describe arbitrary systems," Bach explains. "The premise of computer science, which is experimentally well-validated, is that we can describe any function using state transitions." Joscha Bach
The cognitive scientist argues that these state transitions can be mapped through computational operators that identify discernible differences with finite resolution. This mathematical approach, he contends, provides the formal tools needed to describe any behavior—including consciousness itself.
Bridging Mathematics and Philosophy
Bach identifies a fundamental disconnect between mathematical precision and philosophical inquiry that has historically impeded progress in consciousness research.
"In philosophy, you try to talk about our existence and reality, and the world that we are in, and it's very difficult for those areas to meet mathematical languages. Mathematical languages are so simple that they cannot actually talk about the real world." The AI researcher
The solution to this impasse, according to the Liquid AI scientist, requires a specific mathematical target: "This idea of making philosophy and mathematics meet requires that we are mathematizing the human mind as an intermediator."
Limitations of Human Understanding
Bach doesn't shy away from addressing the challenges inherent in this ambitious project, acknowledging intrinsic limitations in how humans conceptualize their own cognition.
"The human mind itself is limited in our ability to understand because it's quite mushy. We have these vague references grounded in perceptual constructs that we largely only know through pattern matching and not as deep understanding." Bach
This "mushiness" of human cognition, as Bach describes it, explains the difficulty philosophers face when attempting to define concepts like understanding or consciousness with the precision needed for computational implementation.
The computational researcher critiques philosophical approaches that claim "machines cannot understand something" while simultaneously being unable to define what understanding entails. "They just have this very weird pattern matching, and this pattern matching is not good enough to tell them if machines understand something or don't," Bach argues.
The Path Forward
As Bach establishes a new Consciousness Research Institute at Liquid AI, his work aims to close the gap between current machine learning capabilities and human cognition through precise mathematical formalization of mental processes.
By identifying the computational operators and state transitions that underlie consciousness, Bach believes we can develop formal models that capture the essence of subjective experience—potentially revolutionizing both AI development and our understanding of human consciousness.
