# Sparse Overlapping Sets Lasso for Multitask Learning and its Application to fMRI Analysis

arXiv: https://arxiv.org/abs/1311.5422 · HTML: https://mlsanity.com/paper/1311.5422/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Nov 2013.

## Built on by (most cited first, 4 of 4)

- [Scalable Feature Selection for (Multitask) Gradient Boosted Trees](https://mlsanity.com/paper/2109.01965/) (2021)
- [Selective Inference for Sparse Multitask Regression with Applications in Neuroimaging](https://mlsanity.com/paper/2205.14220/) (2022)
- [Universal Feature Selection for Simultaneous Interpretability of Multitask Datasets](https://mlsanity.com/paper/2403.14466/) (2024)
- [SpaRG: Sparsely Reconstructed Graphs for Generalizable fMRI Analysis](https://mlsanity.com/paper/2410.07201/) (2024)

## Abstract (arXiv)

Multitask learning can be effective when features useful in one task are also useful for other tasks, and the group lasso is a standard method for selecting a common subset of features. In this paper, we are interested in a less restrictive form of multitask learning, wherein (1) the available features can be organized into subsets according to a notion of similarity and (2) features useful in one task are similar, but not necessarily identical, to the features best suited for other tasks. The main contribution of this paper is a new procedure called Sparse Overlapping Sets (SOS) lasso, a convex optimization that automatically selects similar features for related learning tasks. Error bounds are derived for SOSlasso and its consistency is established for squared error loss. In particular, SOSlasso is motivated by multi- subject fMRI studies in which functional activity is classified using brain voxels as features. Experiments with real and synthetic data demonstrate the advantages of SOSlasso compared to the lasso and group lasso.