English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Catastrophic forgetting undermines the effectiveness of deep neural networks(DNNs) in scenarios such as continual learning and lifelong learning. Whileseveral methods have been proposed to tackle this problem, there is limitedwork explaining why these methods work well. This paper has the goal of betterexplaining a popularly used technique for avoiding catastrophic forgetting:quadratic regularization. We show that quadratic regularizers preventforgetting of past tasks by interpolating current and previous values of modelparameters at every training iteration. Over multiple training iterations, thisinterpolation operation reduces the learning rates of more important modelparameters, thereby minimizing their movement. Our analysis also reveals twodrawbacks of quadratic regularization: (a) dependence of parameterinterpolation on training hyperparameters, which often leads to traininginstability and (b) assignment of lower importance to deeper layers, which aregenerally the place forgetting occurs in DNNs. Via a simple modification to theorder of operations, we show these drawbacks can be easily avoided, resultingin 6.2\% higher average accuracy at 4.5\% lower average forgetting. We confirmthe robustness of our results by training over 2000 models in differentsettings. Code available at \url{https://github.com/EkdeepSLubana/QRforgetting}

How do Quadratic Regularizers Prevent Catastrophic Forgetting: The Role of Interpolation