Categorical Semantics of Learning by Gradient Descent

Add to your list(s) Download to your calendar using vCal

• Fabio Zanasi (UCL)
• Friday 19 May 2023, 14:00-14:50
• LG23, Computer Science // Zoom.

If you have a question about this talk, please contact George Kaye.

Zoom details

• Link: https://bham-ac-uk.zoom.us/j/81873335084?pwd=T1NaUFg2U1l6d0RLL2RlTzFBam1IUT09
• Meeting ID: 818 7333 5084
• Passcode: 217

Abstract

I will present recent and ongoing work on giving a semantic foundation to training algorithms in machine learning using the categorical formalisms of lenses. Lenses provide a unifying perspective on various classes of such algorithms, as well as offering a different style of specifying and proving properties of training protocols. They also enable the study of machine learning for new classes of models such as Boolean circuits and polynomial circuits. In the last part of the talk I will discuss some applications and directions for future work, including circuit design for hardware implementation, formal verification of learning algorithms, and accounting for learning beyond stochastic gradient descent.

This talk is part of the Theoretical computer science seminar series.

This talk is included in these lists:

• Computer Science Departmental Series
• Computer Science Distinguished Seminars
• LG23, Computer Science // Zoom
• Theoretical computer science seminar
• computer sience

Note that ex-directory lists are not shown.