タイトル | Sparse Bayesian Logistic Regression with Hierarchical Prior and Variational Inference |
---|---|
著者 | 堀井俊佑 |
年度 | 2018 |
形式 | 国際学会 |
分野 | 知識情報処理 |
掲載雑誌名 | Advances in Approximate Bayesian Inference (AABI) 2017 -- NIPS Workshop(ポスター発表) |
掲載号・ページ | |
掲載年 | 2017 |
掲載月 | 12 |
アブスト (日本語) |
Advances in Approximate Bayesian Inference (AABI) 2017 -- NIPS Workshop 2017年12月8日 Seaside Ballroom, Long Beach Convention Center, Long Beach, USA 査読有 DOI:なし http://approximateinference.org/2017/accepted/ |
アブスト (英語) |
In this paper, we present a hierarchical model which assumes the logistic regression function as the observation model and assumes hierarchical priors which promote sparsity of the estimated parameters. We also develop an inference algorithm based on the variational method. The effectiveness of the proposed algorithm is validated through some experiments on both synthetic and real-world data. |
備考 (日本語) |
2 |
備考 (英語) |
2 |
論文原稿 | |
発表資料 |