Abstract | ||
---|---|---|
In many security and healthcare systems, a sequence of sensors/tests is used for detection and diagnosis. Each test outputs a prediction of the latent state, and carries with it inherent costs. Our objective is to learn strategies for selecting a test that gives the best trade-off between accuracy u0026 costs. Unfortunately, it is often impossible to acquire ground truth annotations and we are left with the problem of unsupervised sensor selection (USS). Hanawal et al. [9] reduces USS to a special case of multi-armed bandit problem with side information and develop polynomial time algorithms that achieve sub-linear regret. In this paper, we extend earlier analysis with contextual information, propose an algorithm having sub-linear regret and verify our results on synthetic u0026 real datasets. |
Year | Venue | Field |
---|---|---|
2018 | COMAD/CODS | Contextual information,Regret,Computer science,Side information,Ground truth,Unsupervised learning,Artificial intelligence,Time complexity,Healthcare system,Machine learning,Special case |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
7 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Arun Verma | 1 | 0 | 0.34 |
Manjesh Kumar Hanawal | 2 | 99 | 21.89 |