On Scalability of Active Learning for Formulating Query Concepts
Wei-Cheng Lai, VIMA Technologies and University of California, Santa Barbara
Kingshy Goh, VIMA Technologies and University of California, Santa Barbara
Edward Y. Chang, VIMA Technologies and University of California, Santa Barbara
Query-by-example and query-by-keyword both suffer from the problem of "aliasing," meaning that example-images and keywords potentially have variable interpretations or multiple semantics. For discerning which semantic is appropriate for a given query, we have established that combining active learning with kernel methods is a very effective approach. In this work, we first examine active-learning strategies, and then focus on addressing the challenges of two scalability issues: scalability in dataset size and in concept complexity. We present remedies, explain limitations, and discuss future directions that research might take.