Text Classification with Small Datasets using Deep Transfer Learning
Massive quantities of domain specific labeled data have been the fuel, but also the primary bottleneck for using deep learning algorithms in industry. Organizations which lack the budget, time or have data privacy issues face hurdles in collecting such large amounts of domain specific human annotated data. I will review and compare methods to tackle this problem for text classification tasks via transfer learning using deep learning models. The models discussed will include Universal Sentence Encoders, ELMo and BERT. I will describe the model architectures used, specifics of training mechanisms, the evaluation criteria that guided the experiments and finally provide an attribution analysis of which components contributed most to end performance results.
Hanoz Bhathena is a Data Scientist within the Evidence Lab Innovations division at UBS, where he is responsible for developing machine learning models that uncover insights relevant to investment research. He has experience executing and leading AI/ML projects, particularly those that utilize deep learning, with applications focused on natural language understanding, recommendation systems and search. He holds a Master’s degree in Operations Research from Columbia University and a Bachelor’s degree in Electrical Engineering from the University of Mumbai, VJTI. He has also completed the Artificial Intelligence Graduate Certificate program from Stanford University.