Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the fundamentals of neurosymbolic architectures and their training in this 34-minute tutorial presented by Yisong Yue (Caltech), Swarat Chaudhuri (UT Austin), and Jennifer Sun (Caltech). Begin with an overview of conventional deep learning, focusing on purely neural architectures. Delve into the concept of domain-specific languages (DSLs) that incorporate both symbolic and neural primitives, understanding how these can be used to design programs and architectures. Examine the construction of explicit neurosymbolic architectures and learn how to train their continuous parameters using gradient-based learning techniques. This tutorial serves as an essential introduction to the emerging field of neurosymbolic programming for science, providing a solid foundation for understanding the integration of neural and symbolic approaches in artificial intelligence.
