Overview
This course teaches learners how to efficiently implement distributed deep learning using MXNet. The learning outcomes include understanding practical considerations for machine learning, challenges in deploying large-scale learning, declarative programming, writing parallel programs, and hierarchical parameter servers in MXNet. The course covers topics such as tensors, deep learning, tensor contraction as a layer, Amazon Rekognition for object and scene detection, facial analysis, and Polly for voice quality and pronunciation. The teaching method involves a mix of theoretical concepts and practical examples. This course is intended for individuals interested in distributed deep learning, machine learning practitioners, and developers looking to enhance their skills in implementing deep learning models at scale.
Syllabus
Intro
PRACTICAL CONSIDERATIONS FOR MACHINE LEARNING
CHALLENGES IN DEPLOYING LARGE-SCALE LEARNING
DECLARATIVE PROGRAMMING
MXNET: MIXED PROGRAMMING PARADIGM
WRITING PARALLEL PROGRAMS IS HARD
HIERARCHICAL PARAMETER SERVER IN MXNET
TENSORS, DEEP LEARNING & MXNET
TENSOR CONTRACTION AS A LAYER
Introducing Amazon Al
Rekognition: Object & Scene Detection
Rekognition: Facial Analysis
Polly: A Focus On Voice Quality & Pronunciation
Taught by
Simons Institute