Joseph, Ajin George ; Bhatnagar, Shalabh (2019) An Adaptive and Incremental Approach to Quantile Estimation In: IEEE 58th Conference on Decision and Control (CDC), 11-13 Dec. 2019, Nice, France.
Full text not available from this repository.
Official URL: http://doi.org/10.1109/CDC40024.2019.9029803
Related URL: http://dx.doi.org/10.1109/CDC40024.2019.9029803
Abstract
In this paper, we accomplish two tasks:(1) We provide an alternate proof for the quantile estimation algorithm proposed under a fully relaxed setting. The convergence of the quantile estimation method is guaranteed under two necessary conditions: the stability of the iterates and the differentiability of the expected linear residual function. We literally remove these two assumptions and propose a differential inclusion based analysis. (2) We further propose a much improved quantile estimation algorithm by considering the asymmetrically weighted quadratic residual function along with double averaging.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Source: | Copyright of this article belongs to Institute of Electrical and Electronics Engineers. |
| ID Code: | 116621 |
| Deposited On: | 12 Apr 2021 07:14 |
| Last Modified: | 12 Apr 2021 07:14 |
Repository Staff Only: item control page
