ParallelDots Research & Development

Lifelong Learning algorithm for Medical Image Analysis

In our healthcare product Dentistry.AI, we continually get more and more data annotated to make the prediction better. This is a problem encountered in various Machine Learning based products where one has to take feedback into account to make the algorithm better. We solve this problem of incremental training of AI using example mining. Example Mining has been previously used to address the problem of data imbalance, but this is for the first time someone has proposed to use it for incrementally training AI algorithms on feedback received.

Copyright © ParallelDots,Inc. 2018. All Rights Reserved.