Skip to yearly menu bar Skip to main content


Workshop

Bayesian Incremental Learning for Deep Neural Networks

Maxim Kochurov · Timur Garipov · Dmitrii Podoprikhin · Dmitry Molchanov · Arsenii Ashukha · Dmitry Vetrov

East Meeting Level 8 + 15 #19

Mon 30 Apr, 4:30 p.m. PDT

In industrial machine learning pipelines, data often arrive in parts. Particularly in the case of deep neural networks, it may be too expensive to train the model from scratch each time, so one would rather use a previously learned model and the new data to improve performance. However, deep neural networks are prone to getting stuck in a suboptimal solution when trained on only new data as compared to the full dataset. Our work focuses on a continuous learning setup where the task is always the same and new parts of data arrive sequentially. We apply a Bayesian approach to update the posterior approximation with each new piece of data and find this method to outperform the traditional approach in our experiments.

Live content is unavailable. Log in and register to view live content