Vous êtes sur la page 1sur 26

AI/ML and

Mobile
Hello!
I am Paramvir Singh

I am Android developer
and consultant.

You can find me at:


@paramvirsingh88

2
Agenda

◎ Understand AI and ML
◎ Different tools and frameworks
◎ Using ML in mobile apps - ML Kit

3
1.
Artificial
Intelligence
and Machine Learning

4

Artificial intelligence would be the
ultimate version of Google. The
ultimate search engine that would
understand everything on the web.
—Larry Page

5
AI vs ML

AI ML
A branch of computer A computer program is
science the objective of fed a lot of data and
which is to make learns from experiences.
machine work as smart A subset of AI.
as human.
Cognitive ability of
machines.

6
7
A new
Tech?
Large datasets,
advancements in data
science, powerful
hardware

8
Types of Learning

Supervised Unsupervised Reinforcement


Output is fed into the Success and failure Once result is
system. is not known to the obtained, it is
Labeled data. system. categorized as
No labeling. success or failure.
Pattern recognition. Similar to
unsupervised.

9
Understanding Machine Learning

◎ Convolution matrix
◎ Deep neural network, tensors
◎ Dynamic computation, computational
graphs
◎ Automatic differentiation
◎ Relation Extraction, Coreference Resolution

10
Understanding Machine Learning

◎ Convolution matrix
◎ Deep neural network, tensors
◎ Dynamic computation, computational
graphs
◎ Automatic differentiation
◎ Relation Extraction, Coreference Resolution

11
Instead we will go the easier route

◎ Data
◎ Model
◎ Training
◎ Learning
◎ Experience
◎ Decision

12
How ML models work

Process Data Train Model Apply model

13
How it’s all spun together

● Huge datasets required


● Create and train models
● Choose best model
● Apply the model

14
ML models

15
ML models

16
Applications

● Image processing
● NLP and Speech recognition
● Predictions
○ Diagnostic
○ Business and customers
○ Natural events
● Robotics/real time decisions

17
Some recent examples

■ First AI art
■ Cambridge
Analytica case
■ Cancer detection
■ Self driving cars
■ Google Translate
and email smart
reply

18
ML frameworks

■ Cloud Vision
■ ML kit
■ TensorFlow
■ Amazon Rekognition
■ Facebook Pytorch
■ OpenCV
■ Azure cloud ML

19
ML Kit
Easy Machine
Learning

20
ML Kit

Mobile focused API Based


Younger sibling of No mode creation
Google Cloud or model training
Vision. required.
Uses models
provided and
hosted by Google.
Focussed on Vision
but will expand in
future.

21
What problem does it solve?

■ Doing ML from scratch is doable but hard


■ Huge data required
■ Knowledge of math and ML science required
■ Domain knowledge required

22
What ML Kit provides

On device ML
processing
Model
hosting in
cloud

Firebase
API’s

23
ML Kit features

Image Labeling Text Detection Face detection

Barcode scanning Landmark Smart Reply*


detection

*coming soon 24
Integration in
Android
Integrate Firebase and you are
good to go.

25
Thanks!
Any questions?

You can find me at:


◎ @paramvirsingh88
◎ paramvir.singh88@gmail.com

26

Vous aimerez peut-être aussi