| EPSRC Reference: |
GR/K70366/01 |
| Title: |
THE CANONICAL METRIC IN MACHINE LEARNING: THEORY AND APPLICATIONS |
| Principal Investigator: |
Shawe-Taylor, Professor JS |
| Other Investigators: |
|
| Researcher Co-Investigators: |
|
| Project Partners: |
|
| Department: |
Computer Science |
| Organisation: |
Royal Holloway, Univ of London |
| Scheme: |
Standard Research (Pre-FEC) |
| Starts: |
01 July 1995 |
Ends: |
31 December 1997 |
Value (£): |
46,775
|
| EPSRC Research Topic Classifications: |
| Fundamentals of Computing |
|
|
| EPSRC Industrial Sector Classifications: |
|
| Related Grants: |
|
| Panel History: |
|
|
Summary on Grant Application Form |
|
It is well known that the main difficulty in machine learning is the preliminary biasing of the learner (selecting the appropriate features, finding the right model, etc.). Once suitable bias has been found the actual learning task is considerably easier. At present the preliminary biasing of a learner is an ad hoc procedure with no body of theoretical results or algorithms to aid and formalise the process. Recently, however, the named Research Assistant (Jonathan Baxter) in his Doctoral Dissertation has formulated a new model of machine learning that does include the preliminary biasing phase and has developed algorithms for learning the preliminary bias. Theoretical and experimental results show that learning performance within the new model is greatly improved over classical methods. The most refined formulation of the model to date is in terms of the so-called canonical metric - a distance measure on the input space of the learner, knowledge of which is equivalent to knowledge of the appropriate bias for the learning domain. The purpose of the proposed research is twofold: (1) To further investigate the theory of the canonical metric, in particular sampling bounds for its estimation and techniques for its embedding in Euclidean space; (2) To test the model on the more real-world problems of character and speech cognition.
|
|
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: |
|