# Multitask Learning with Learned Task Relationships

arXiv: https://arxiv.org/abs/2510.10570 · HTML: https://mlsanity.com/paper/2510.10570/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 27 references that are in the map.
- First posted on arXiv 12 Oct 2025.

## Builds on (most cited first, 10 of 10)

- [Federated Learning: Challenges, Methods, and Future Directions](https://mlsanity.com/paper/1908.07873/) (2019)
- [Personalized Federated Learning with Moreau Envelopes](https://mlsanity.com/paper/2006.08848/) (2020)
- [Adaptive Gradient-Based Meta-Learning Methods](https://mlsanity.com/paper/1906.02717/) (2019)
- [Multi-Institutional Deep Learning Modeling Without Sharing Patient Data: A Feasibility Study on Brain Tumor Segmentation](https://mlsanity.com/paper/1810.04304/) (2018)
- [Graph Learning based Recommender Systems: A Review](https://mlsanity.com/paper/2105.06339/) (2021)
- [Dif-MAML: Decentralized Multi-Agent Meta-Learning](https://mlsanity.com/paper/2010.02870/) (2020)
- [Learning over Multitask Graphs -- Part I: Stability Analysis](https://mlsanity.com/paper/1805.08535/) (2018)
- [Multitask Diffusion Adaptation over Networks](https://mlsanity.com/paper/1311.4894/) (2013)
- [Learning over Multitask Graphs -- Part II: Performance Analysis](https://mlsanity.com/paper/1805.08547/) (2018)
- [Distributed Clustering and Learning Over Networks](https://mlsanity.com/paper/1409.6111/) (2014)

## Built on by (most cited first, 1 of 1)

- [Decentralized Multitask Learning over Learned Task Graphs](https://mlsanity.com/paper/2608.26989/) (2026)

## Abstract (arXiv)

Classical consensus-based strategies for federated and decentralized learning are statistically suboptimal in the presence of heterogeneous local data or task distributions. As a result, in recent years, there has been growing interest in multitask or personalized strategies, which allow individual agents to benefit from one another in pursuing locally optimal models without enforcing consensus. Existing strategies require either precise prior knowledge of the underlying task relationships or are fully non-parametric and instead rely on meta-learning or proximal constructions. In this work, we introduce an algorithmic framework that strikes a balance between these extremes. By modeling task relationships through a Gaussian Markov Random Field with an unknown precision matrix, we develop a strategy that jointly learns both the task relationships and the local models, allowing agents to self-organize in a way consistent with their individual data distributions. Our theoretical analysis quantifies the quality of the learned relationship, and our numerical experiments demonstrate its practical effectiveness.