Chief Data Scientist, RoundSqr
Srinivas is the quintessential AI/ML geek, with an innate ability to solve business problems using insights from data. He has over 23 years as an Analytics / Information Management practitioner across the US, UK, and India, in the diverse roles of an Engineer, Architect, Data Analyst & Data Scientist.
WATCH LIVE: November 2nd at 9:30 am
Many Active Learning methods belong to the re-training based approaches, which select one unlabeled instance, add it to the training set with it’s possible labels, retrain the classification model and evaluate the criteria we base our selection on. However, since the true label of the selected instance is unknown, these methods resort to calculating the average-case or worst-case performance wrt the unknown label. The above method aims to make use of the Uncertainty information to enhance the performance of retraining – based models.