Andrew Watts-Curnow, Senior Cloud Architect – Professional Services, APAC, AWS
Learn how advances in AI are enabling improvements in customer experience. This is a deep dive using machine learning frameworks for people who are familiar with building their own models. In this session, we will detail a facial recognition solution that can detect known customers and alert customer service staff.
2. A Long History of AI at Amazon
Personalized
recommendations
Inventing entirely
new customer
experiences
Fulfillment
automation and
inventory
management
Drones Voice driven
interactions
7. Amazon Rekognition Video
Use case: Customer Facial Recognition
Live Store Camera Amazon Kinesis Video Streams Amazon Rekognition Video Face collection
1. Camera-captured video
streams are processed by
Kinesis Video Streams
2. Amazon Rekognition Video analyses
the video and searches faces on screen
against a collection of millions of faces
Staff AWS Lambda Amazon Kinesis
Streams
8. Build your own AI
Platform Services
Application Services
Frameworks & Interfaces
Caffe2 CNTK
Apache
MXNet
PyTorch TensorFlow Torch Keras Gluon
AWS Deep Learning AMIs
Amazon SageMaker AWS DeepLens
Rekognition Transcribe Translate Polly Comprehend Lex
11. AI is still too complicated for everyday developers
Collect and
prepare training
data
Choose and
optimize your
AI algorithm
Set up and
manage
environments
for training
Train and tune
model
(trial and error)
Deploy model
in production
Scale and
manage the
production
environment
13. Amazon SageMaker
Pre-built
notebooks for
common
problems
K-Means Clustering
Principal Component Analysis
Neural Topic Modelling
Factorization Machines
Linear Learner - Regression
XGBoost
Latent Dirichlet Allocation
Image Classification
Seq2Seq
Linear Learner - Classification
ALGORITHMS
Apache MXNet
TensorFlow
Caffe2, CNTK,
PyTorch, Torch
FRAMEWORKS
Set up and manage
environments for
training
Train and tune
model (trial and
error)
Deploy model
in production
Scale and manage the
production environment
Built-in, high
performance
algorithms
BUILD
20. Amazon EC2 P3 Instances
• Up to eight NVIDIA Tesla V100 GPUs
• 1 PetaFLOP of computational performance
• 300 GB/s GPU-to-GPU communication (NVLink)
• 16GB GPU memory with 900 GB/sec peak GPU
memory bandwidth
T h e f a s t e s t , m o s t p o w e r f u l G P U i n s t a n c e s i n t h e c l o u d
22. Hyper Parameter Optimization
• Discover a model’s optimal hyper parameters.
• Optimize for accuracy, precision, f1, training time, etc.
• Meta-model regression of learning rate, dropoff,
layers, etc.
23. Amazon SageMaker
Fully managed
hosting with auto-
scaling
One-click
deployment
Pre-built
notebooks for
common
problems
Built-in, high
performance
algorithms
One-click
training
Hyperparameter
optimization
BUILD TRAIN DEPLOY
25. AWS DeepLens
HD video camera
Custom-designed
deep learning
inference engine
Micro-SD
Mini-HDMI
USB
USB
Reset
Audio out
Power
HD video camera
with on-board
compute optimized
for deep learning
Tutorials, examples,
demos, and pre-built
models
From unboxing to
first inference in
<10 minutes
Integrates with Amazon
SageMaker and AWS
Lambda
10
MIN