Adam KMulti-task learning: what is it, how does it work and why does it work?One of the most exciting and seemingly ubiquitous recent topics in deep learning (DL) is without a doubt multi-task learning. Multi-task…Nov 11, 20222Nov 11, 20222
Ajit SarkaarDomain Apposite Pre-processing to Improve Classification PerformanceWe will discuss reasons why widely adopted rescaling conventions may not give the best results for our scene categorization task.Apr 28, 2022Apr 28, 2022
Sanja WidmerTaking a close look at Precision for classification tasksAt GumGum, we train machine learning models on unstructured data, like text and images. An integral part of model development is model…Nov 24, 2021Nov 24, 2021
Sanja WidmerTaking a fresh look at metrics for Classification Tasks at GumGum — IntroductionAt GumGum we train machine learning models on unstructured data, like text and images. An integral part of model development is model…Oct 28, 2021Oct 28, 2021
Sooraj SubrahmannianContextual Brand Safety -IIContextual brand safety is an ongoing series. We describe the steps taken to train a multi-label text classifier for brand safetyMay 19, 2020May 19, 2020