Multi-Task Learning — Technology Wiki

Overview

Direct Answer

Multi-task learning is a machine learning paradigm in which a single model is trained simultaneously on multiple related prediction tasks, leveraging shared representations to improve generalisation and reduce overfitting compared to training separate task-specific models.

How It Works

The model architecture contains shared hidden layers that extract common features, with task-specific output heads branching from these layers. During training, loss functions from all tasks are combined—typically through weighted summation—and gradients propagate through both task-specific and shared parameters, forcing the model to learn representations beneficial across tasks.

Why It Matters

Multi-task learning reduces data requirements per task, decreases training time and computational cost, and improves robustness by allowing models to exploit task relationships. Organisations benefit from fewer model deployments and improved performance in data-constrained domains such as rare disease diagnosis or low-resource languages.

Common Applications

Applications include natural language processing (simultaneous part-of-speech tagging, named entity recognition, and dependency parsing), computer vision (joint detection, segmentation, and classification), and autonomous systems (predicting vehicle trajectory alongside traffic sign recognition).

Key Considerations

Effective multi-task learning requires careful selection of related tasks; incompatible or weakly related tasks can degrade performance through negative transfer. Task weighting and architectural design significantly influence outcomes and often require domain expertise to optimise.

Cross-References(1)

Machine Learning

Referenced By1 term mentions Multi-Task Learning

Other entries in the wiki whose definition references Multi-Task Learning — useful for understanding how this concept connects across Machine Learning and adjacent domains.

Catastrophic Forgetting·Machine Learning

Related in MLOps & Production

Machine Learning

A subset of AI that enables systems to automatically learn and improve from experience without being explicitly programmed.

Supervised Learning

A machine learning paradigm where models are trained on labelled data, learning to map inputs to known outputs.

Unsupervised Learning

A machine learning approach where models discover patterns and structures in data without labelled examples.

Reinforcement Learning

A machine learning paradigm where agents learn optimal behaviour through trial and error, receiving rewards or penalties.

Online Learning

A machine learning method where models are incrementally updated as new data arrives, rather than being trained in batch.

Batch Learning

Training a machine learning model on the entire dataset at once before deployment, as opposed to incremental updates.

Active Learning

A machine learning approach where the algorithm interactively queries a user or oracle to label new data points.

Ensemble Learning

Combining multiple machine learning models to produce better predictive performance than any single model.

Feature Selection

The process of identifying and selecting the most relevant input variables for a machine learning model.

Epoch

One complete pass through the entire training dataset during the machine learning model training process.

Model Serialisation

The process of converting a trained model into a format that can be stored, transferred, and later reconstructed for inference.

Model Serving

The infrastructure and processes for deploying trained machine learning models to production environments for real-time predictions.

More in Machine Learning

t-SNE

Unsupervised Learning

t-Distributed Stochastic Neighbour Embedding — a technique for visualising high-dimensional data in two or three dimensions.

Catastrophic Forgetting

Anomaly & Pattern Detection

The tendency of neural networks to completely lose previously learned knowledge when trained on new tasks, a fundamental challenge in continual and multi-task learning.

Data Augmentation

Feature Engineering & Selection

Techniques that artificially increase the size and diversity of training data through transformations like rotation, flipping, and cropping.

Model Registry

MLOps & Production

A versioned catalogue of trained machine learning models with metadata, lineage, and approval workflows, enabling reproducible deployment and governance at enterprise scale.

Decision Tree

Supervised Learning

A tree-structured model where internal nodes represent feature tests, branches represent outcomes, and leaves represent predictions.

Gradient Boosting

Supervised Learning

An ensemble technique that builds models sequentially, with each new model correcting residual errors of the combined ensemble.

Markov Decision Process

Reinforcement Learning

A mathematical framework for modelling sequential decision-making where outcomes are partly random and partly controlled.

Naive Bayes

Supervised Learning

A probabilistic classifier based on applying Bayes' theorem with the assumption of independence between features.