WEBVTT

00:00:18 --> 00:02:53
Machine learning opportunities within audience interaction platforms

00:02:53 --> 00:05:12
Extracting key phrases and analyzing sentiment from audience questions

00:05:12 --> 00:08:53
Detecting similar questions and generating automated quiz responses

00:08:53 --> 00:11:45
Moving to serverless deployments for Python machine learning models

00:11:45 --> 00:14:24
Addressing constraints and limitations of serverless model deployments

00:14:24 --> 00:17:50
Fine-tuning BERT models for audience sentiment analysis classification

00:17:50 --> 00:22:24
Reducing model size using knowledge distillation with teacher-student models

00:22:24 --> 00:26:22
Optimizing similar question detection with linear complexity sentence encoders

00:26:22 --> 00:30:34
Distilling cross-encoder models into smaller efficient sentence embedding models

00:30:34 --> 00:35:16
Replacing PyTorch with ONNX runtime for AWS Lambda deployments

00:35:16 --> 99:59:59
Evaluating cost effectiveness and summarizing serverless machine learning strategies