Research map / Learning theory and algorithms / Matrix and tensor methods
Optimal Transfer Learning for Missing Not-at-Random Matrix Completion
Akhil Jalan, Yassir Jedra, Arya Mazumdar, Soumendu Sundar Mukherjee, Purnamrita Sarkar
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Studies transfer learning for MNAR matrix completion, using a noisy incomplete source matrix related by latent feature shift to help estimate a target matrix with missing rows and columns, under active and passive sampling.
Problem
Matrix completion when the target matrix has entire rows and columns missing not at random, making estimation impossible without side information; motivated by biological problems.
Approach
Uses a noisy, incomplete source matrix linked to the target by a feature shift in latent space to transfer information, with both active and passive sampling designs analyzed for optimal transfer.
Closest papers by meaning
- A Pairwise Pseudo-likelihood Approach for Matrix Completion with Informative Missingness (NeurIPS 2024 · similarity 0.89)
- A New Theory for Matrix Completion (NeurIPS 2017 · similarity 0.88)
- Noise-Tolerant Life-Long Matrix Completion via Adaptive Sampling (NeurIPS 2016 · similarity 0.88)
- High Rank Matrix Completion With Side Information (AAAI 2018 · similarity 0.88)
- Scalable Completion of Nonnegative Matrix with Separable Structure (AAAI 2016 · similarity 0.87)
- Symmetric Matrix Completion with ReLU Sampling (ICML 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
