| EPSRC Reference: |
GR/R46670/01 |
| Title: |
Complexity Approximation Principle and Predictive Complexity: Analysis and Applications |
| Principal Investigator: |
Gammerman, Professor A |
| Other Investigators: |
|
| Researcher Co-Investigators: |
|
| Project Partners: |
|
| Department: |
Computer Science |
| Organisation: |
Royal Holloway, Univ of London |
| Scheme: |
Standard Research (Pre-FEC) |
| Starts: |
01 October 2001 |
Ends: |
31 December 2004 |
Value (£): |
142,996
|
| EPSRC Research Topic Classifications: |
| Artificial Intelligence |
Fundamentals of Computing |
|
| EPSRC Industrial Sector Classifications: |
| Communications |
Pharmaceuticals and Biotechnology |
| Information Technologies |
|
|
| Related Grants: |
|
| Panel History: |
|
|
Summary on Grant Application Form |
|
The research will concentrate on properties of predictive complexity and the Complexity Approximation Principle, which is an application of predictive complexity to model selection.Predictive complexity is a generalisation of Kolmogorov complexity motivated by learning theory; it bounds the ability of an algorithm to predict the elements of a sequence and thus sets the limits for machine learning. It is proposed to investigate which properties of Kolmogorov complexity can be generalised for predictive complexity and to what extent. In particular, the notions of conditional complexity and information will be extended to the case of predictive complexity and the concept of randomness will be studied. This research is expected to provide a new insight to the problem of finding dependences between processes.Complexity Approximation Principle generalises the well-known Minimum Description Length and Minimum Message Length principles. It transforms the idea to balance the goodnexx-of-fit of a hypothesis against its complexity to a wider class of learning settings. It is planned to implement Complexity Approximation Principle to particular learning problems, to develop the corresponding learning algorithms, and to compare their performance with other learning methods including other regularisation techniques.
|
|
Key Findings |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
|
Potential use in non-academic contexts |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
|
Impacts |
|
| Description |
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk |
| Summary |
|
| Date Materialised |
|
|
|
|
Sectors submitted by the Researcher |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
| Project URL: |
|
| Further Information: |
|
| Organisation Website: |
|