Abstract | ||
---|---|---|
In many estimation problems, the measure- ment process can be actively controlled to alter the information received. The control choices made in turn determine the perfor- mance that is possible in the underlying in- ference task. In this paper, we discuss perfor- mance guarantees for heuristic algorithms for adaptive measurement selection in sequential estimation problems, where the inference cri- terion is mutual information. We also demon- strate the performance of our tighter online computable performance guarantees through computational simulations. |
Year | Venue | Keywords |
---|---|---|
2007 | AISTATS | computer simulation,heuristic algorithm,mutual information,sequential estimation |
Field | DocType | Volume |
Mathematical optimization,Heuristic,Computer science,Inference,Mutual information,Artificial intelligence,Sequential estimation,Machine learning | Journal | 2 |
Citations | PageRank | References |
6 | 0.57 | 7 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jason L. Williams | 1 | 217 | 15.34 |
John W. Fisher III | 2 | 878 | 74.44 |
Alan S. Willsky | 3 | 7466 | 847.01 |