A Fast Projection onto the Simplex and the l 1 Ball

By Laurent Condat. 2015

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Ben Peters
2. Vlad Niculae
3. André F. T. Martins
4. Cem Anil
5. James Lucas
6. Roger Baker Grosse
7. Yoshua Bengio
8. Ruslan Salakhutdinov

Citations in the corpus

• Sorting out Lipschitz function approximation (2018, methodology): “Due to its simplicity and suitability for a GPU implementation, we use Algorithm 1 from Condat [13] to project the weight matrices onto the L ∞ ball in all of our experiments.”
• Sparse Sequence-to-Sequence Models (2019, methodology): “Further speed-ups might be available following the strategy of Condat (2016), but our simple incremental method is very easy to implement on the GPU using primitives available in popular libraries (Paszke et al., 2017).”