# Residual Correlation as a Diagnostic for Joint-Uncertainty Gains from GP Coregionalisation

arXiv: https://arxiv.org/abs/2609.30085 · HTML: https://mlsanity.com/paper/2609.30085/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 39 references that are in the map.
- First posted on arXiv 24 Sep 2026.

## Builds on (most cited first, 17 of 17)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [A Survey on Multi-Task Learning](https://mlsanity.com/paper/1707.08114/) (2017)
- [An Overview of Multi-Task Learning in Deep Neural Networks](https://mlsanity.com/paper/1706.05098/) (2017)
- [Accurate Uncertainties for Deep Learning Using Calibrated Regression](https://mlsanity.com/paper/1807.00263/) (2018)
- [Which Tasks Should Be Learned Together in Multi-task Learning?](https://mlsanity.com/paper/1905.07553/) (2019)
- [When Gaussian Process Meets Big Data: A Review of Scalable GPs](https://mlsanity.com/paper/1807.01065/) (2018)
- [LEEP: A New Measure to Evaluate Transferability of Learned Representations](https://mlsanity.com/paper/2002.12462/) (2020)
- [LogME: Practical Assessment of Pre-trained Models for Transfer Learning](https://mlsanity.com/paper/2102.11005/) (2021)
- [An Information-Theoretic Approach to Transferability in Task Transfer Learning](https://mlsanity.com/paper/2212.10082/) (2022)
- [Frustratingly Easy Transferability Estimation](https://mlsanity.com/paper/2106.09362/) (2021)
- [Heterogeneous Multi-output Gaussian Process Prediction](https://mlsanity.com/paper/1805.07633/) (2018)
- [Generalized Multi-Output Gaussian Process Censored Regression](https://mlsanity.com/paper/2009.04822/) (2020)
- [On Negative Transfer and Structure of Latent Functions in Multi-output Gaussian Processes](https://mlsanity.com/paper/2004.02382/) (2020)
- [Fully Heteroscedastic Count Regression with Deep Double Poisson Networks](https://mlsanity.com/paper/2406.09262/) (2024)
- [Multi-Target Tobit Models for Completing Water Quality Data](https://mlsanity.com/paper/2302.10648/) (2023)
- [Non-stationary and Sparsely-correlated Multi-output Gaussian Process with Spike-and-Slab Prior](https://mlsanity.com/paper/2409.03149/) (2024)

## Abstract (arXiv)

In multi-target regression, correlated targets are often coupled through multi-output Gaussian processes with an intrinsic model of coregionalisation (GP-ICM), assuming that sharing statistical strength improves overall performance. In practice, the benefits are inconsistent. Across the settings studied, we find that the main benefit of coregionalisation is joint uncertainty quantification rather than point prediction. Raw target correlation does not predict when coupling helps; in the separable GP-ICM settings studied here, residual correlation, the cross-target dependence left unexplained by independent per-target predictors, is the strongest predictor of joint-uncertainty gains. We introduce a lightweight diagnostic, Drm logdet=-frac12logdet Rrm res, which represents the idealised joint negative log-likelihood (NLL) gain from modelling a full rather than diagonal residual covariance and is computable from independent GPs alone. Across a controlled synthetic study, 16 multi-target benchmarks, and frozen transformer and convolutional neural network representations for keypoint regression, point prediction remains largely unchanged (Delta R2≈ 0). In contrast, Drm logdet strongly predicts observed ICM NLL improvements (ρs=-0.83, p<0.001), outperforming heuristics such as the feature-to-sample ratio. We also propose Residual-ICM, which preserves independent marginal variances while adding residual-correlation structure to the joint covariance. Residual-ICM achieves the best average joint NLL among the compared methods, while the diagnostic indicates when covariance coupling is likely to be useful. The diagnostic is specific to global Gaussian residual dependence, the structure captured by separable coregionalisation.