Guillaume Dalle
Interventions
02/10
08:50
10minutes
Opening
Guillaume Dalle
-
General
Amphithéâtre Robert Faure
03/10
11:30
30minutes
Leveraging Sparsity to Accelerate Automatic Differentiation
Adrian Hill, Guillaume Dalle
Jacobians and Hessians play vital roles in scientific computing and machine learning, from optimization to probabilistic modeling. While these matrices are often considered too computationally expensive to calculate, their inherent sparsity can be leveraged to dramatically accelerate Automatic Differentiation (AD). By building on top of DifferentiationInterface.jl, we are able to bring Automatic Sparse Differentiation to all major Julia AD backends, including ForwardDiff and Enzyme.
Error, derivatives, stability
Amphithéâtre Jean-Baptiste Say
03/10
12:40
10minutes
Closing
Guillaume Dalle
-
General
Amphithéâtre Robert Faure