Using MATLAB®, engineers and other domain experts have deployed thousands of machine learning applications. MATLAB makes the hard parts of machine learning easy with:
- Point-and-click apps for training and comparing models
- Advanced signal processing and feature extraction techniques
- Automatic machine learning (AutoML) including feature selection, model selection and hyperparameter tuning
- The ability to use the same code to scale processing to big data and clusters
- Automated generation of C/C++ code for embedded and high-performance applications
- Integration with Simulink as native or MATLAB Function blocks, for embedded deployment or simulations
- All popular classification, regression, and clustering algorithms for supervised and unsupervised learning
- Faster execution than open source on most statistical and machine learning computations
See How Others Use MATLAB for Machine Learning
PathPartner Develops Machine Learning Algorithms for Radar-Based Automotive Applications
Energy Production and IA&M
RWE Renewables, Hydro Quebec, IMCORP
Utility Asset Condition Monitoring and Predictive Maintenance using Machine Learning and Artificial Intelligence
Medical Devices and CES
Kinesis Health Technologies
Assessing the Risk of Falls in Older Adults with Inertial Sensors and Machine Learning
Choose from a wide variety of the most popular classification, clustering, and regression algorithms – now also “shallow” neural nets (up to three layers) alongside other machine learning models. Use classification and regression apps to interactively train, compare, tune, and export models for further analysis, integration, and deployment. If writing code is more your style, you can further optimize models with feature selection and parameter tuning.
Overcome the black-box nature of machine learning by applying established interpretability methods such as Partial Dependence plots, LIME, Shapley values, and Generalized Additive Model (GAM). Validate that the model is using the right evidence for its predictions and find model biases that were not apparent during training.
Automatically generate features from training data and optimize models using hyperparameter tuning techniques such as Bayesian optimization. Use specialized feature extraction techniques such as wavelet scattering for signal or image data, and feature selection techniques such as neighborhood component analysis (NCA), minimum redundancy maximum relevance (MRMR) or sequential feature selection.
Deploy statistics and machine learning models to embedded systems and generate readable C or C++ code for your entire machine learning algorithm, including pre and post processing steps. Accelerate verification and validation of your high-fidelity simulations using machine learning models through MATLAB function blocks and native blocks in Simulink.
Scaling & Performance
Use tall arrays train machine learning models to data sets too large to fit in memory, with minimal changes to your code. You can also speed up statistical computations and model training with parallel computing on your desktop, on clusters, or on the cloud.