University of Cambridge > Talks.cam > Wednesday Seminars - Department of Computer Science and Technology  > Hard and soft equivariance priors via Steerable CNNs

Hard and soft equivariance priors via Steerable CNNs

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

If you have a question about this talk, please contact Ben Karniely.

Equivariance can enhance the data efficiency of machine learning models by incorporating prior knowledge about a problem. Thanks to their flexibility and generality, steerable CNNs are a popular design choice for equivariant networks. By leveraging concepts from harmonic analysis, these networks model symmetries through specific constraints on their learnable weights or filters. This framework facilitates the practical implementation of a wide variety of equivariant architectures – e.g. to most Euclidean isometries, including E(3), E(2) and their subgroups.

However, unknown or imperfect symmetries can sometimes lead to overconstrained weights and suboptimal performance. This challenge motivated the study of strategies to enforce softer priors into the models. In the second half of this talk, we will discuss a novel probabilistic approach to learning the degrees of equivariance in steerable CNNs. The method replaces the equivariance constraint on the weights with an expectation over a learnable distribution, which is analytically computed by leveraging its Fourier decomposition.

Link to join virtually: https://cam-ac-uk.zoom.us/j/87421957265

This talk is being recorded. If you do not wish to be seen in the recording, please avoid sitting in the front three rows of seats in the lecture theatre. Any questions asked will also be included in the recording. The recording will be made available on the Department’s webpage

This talk is part of the Wednesday Seminars - Department of Computer Science and Technology series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2025 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity