Overview
The Machine Learning Group specializes in developing algorithms for automatic pattern recognition, prediction, analysis, classification, and learning of structures. We supply both core technologies and machine learning services.
Our core technologies include:
- Bayesian networks
- Learning and classifying structures
- Anomaly detection
- Feature selection
- Time series analysis
- Support Vector Machines (SVM)
Our current activities include:
- CLERK - Clustering with ExteRnal Knowledgebase
- AdAd - Addressable Advertizing
- HYPERGENES
- Anatomic and Symbolic Mapper Engine
- EuResist
- PML - Parallel Machine Learning
- IR - Information Retrieval
- 300mm
- MeLoDy - Machine Learning for Dynamic System Analysis
- Vigilant (virtual guest inspection, learning, and control)
- MilePost - MachIne Learning for Embedded PrOgramS opTimization
- CDG - Coverage Directed Test Generation
- CodeGuru
In the course of our activities, we developed several tools to facilitate our work. These include:
- Bayesian networks toolbox
- Pattern classification toolbox for Matlab
- Parallel SVM implementation
- Client and server application for running parallel algorithms in Matlab
We develop and support simulation-based verification tools, including:
- Coverage directed test generation for functional verification
- Probabilistic regression suites for functional verification
- Harnessing machine learning to improve the success rate of test generation
Our team provides cross-departmental machine learning services, for areas such as:
- Information Retrieval: Enhancing the handling of difficult queries
- Bioinformatics and biodata mining: Peptide identification, prediction of responses to anti-HIV treatment
- Time series analysis, prediction and anomaly detection: Distributed storage systems, call centers
- Projects for companies outside IBM
- Autonomic Computing: Identification of faults in computers
