machine learning simplified€¦ · consider machine learning when update as more data becomes...
TRANSCRIPT
![Page 1: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/1.jpg)
1© 2015 The MathWorks, Inc.
Machine Learning Simplified
Paola Jaramillo
![Page 3: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/3.jpg)
3
New MATLAB framework makes machine learning
easy and accessible for Engineers
![Page 4: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/4.jpg)
4
Consider Machine Learning When
update as more data becomes available
learn complex non-linear relationships
learn efficiently from very large data sets
Solution is too complex for hand written rules or equations
Speech Recognition Object Recognition Engine Health Monitoring
Solution needs to adapt with changing data
Weather Forecasting Energy Load Forecasting Stock Market Prediction
Solution needs to scale
IoT Analytics Taxi Availability Airline Flight Delays
Because algorithms can
![Page 5: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/5.jpg)
5
Dogs
…Dog
Car
Cat
Bird
What is Machine Learning?
Machine learning algorithms use computational methods to “learn” information
directly from data without assuming a predetermined equation as a model
Extract Features Train ModelsTraining data
94%
![Page 6: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/6.jpg)
6
Challenges
Domain-
Expertise
Access
Data
Extract
Features
Share
Models
Develop
Models
Software
Engineering
Data-
Science
![Page 7: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/7.jpg)
7
Challenges from our Customers
Goal: Develop a predictive maintenance system to reduce
pump equipment costs and downtime.
• Convert unreadable data into a usable format.
• Automate filtering, spectral analysis, and
transform steps for multiple trucks and regions.
Goal: Develop a prototype quickly, relying on functions
that have been deployed across ASML’s large, diverse
user base and maintained by dedicated professionals.
• Lack of experience with neural networks or
machine learning.
![Page 8: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/8.jpg)
8
New MATLAB framework makes machine learning
easy and accessible for Engineers
![Page 9: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/9.jpg)
9
Access Data Extract Features Develop Models Integrate
MATLAB makes Machine Learning Easy and Accessible…
… with industry
proven solutions
… from idea
to product
… enabling non-
experts
![Page 10: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/10.jpg)
10
Using Machine Learningto build and deploy a predictive maintenance system
Predictive Model
deployed to drill site
Pump logs
of temperature, pressure
& other data
1TB
Analytics and
Machine Learning
plus signal processing,
neural networks & more
Maintenance
Needed
![Page 11: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/11.jpg)
11
Autonomous Braking System, Scania
![Page 12: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/12.jpg)
12
Our Customers Achievements
“MATLAB gave us the ability to convert previously unreadable
data into a usable format; automate filtering, spectral analysis, and
transform steps for multiple trucks and regions; and ultimately, apply
machine learning techniques in real time to predict the ideal
time to perform maintenance.”
Gulshan Singh
Baker Hughes
“As a process engineer I had no experience with neural networks or
machine learning. I worked through the MATLAB examples to find the best
machine learning functions for generating virtual metrology. I couldn’t
have done this in C or Python—it would’ve taken too long to
find, validate, and integrate the right packages.”
Emil Schmitt-Weaver
ASML
![Page 13: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/13.jpg)
13
Summary of Results
• Savings of more than $10 million projected
• Development time reduced tenfold
• Multiple types of data easily accessed
• Industry leadership established
• Potential manufacturing improvements identified
• Maintenance overhead minimized
![Page 14: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/14.jpg)
14
Artificial Intelligence
Machine Learning
Artificial Intelligence, Machine Learning and Deep Learning
Deep Learning
Timeline
1950s Today1980s
Applic
ation B
readth
Automated Driving
Speech Recognition
RoboticsObject Recognition
Bioinformatics
Recommender Systems
Spam Detection
Fraud Detection
Weather Forecasting
Algorithmic Trading
Sentiment Analysis
Medical Diagnosis
Health Monitoring
Computer Board Games
Machine Translation
Knowledge Representation
Perception
Reasoning
Interactive Programs
Expert Systems
![Page 15: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/15.jpg)
15
Cat
Dog
Bird
Car
Learned Features
…
Car
Dog
Cat
Bird
What is Deep Learning?
Deep learning is a type of machine learning that learns tasks
directly from data
![Page 16: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/16.jpg)
16
Why is Deep Learning So Popular Now?
Source: ILSVRC Top-5 Error on ImageNet
Human
Accuracy
![Page 17: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/17.jpg)
17
Deep Learning Enablers
Massive sets of labeled data
Acceleration with GPUs
Availability of state of the art models from experts
![Page 18: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/18.jpg)
18
MATLAB makes Deep Learning Easy and Accessible
• Handle large images sets
• Accelerate with GPUs
• Visualize and debug networks
• Access pre-trained models
![Page 19: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/19.jpg)
19
Making Deep Learning easy to use is Changing the World
![Page 20: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/20.jpg)
20
Training & Consulting
Public
On-Site
Data processing
Machine Learning
Computer Vision
![Page 21: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/21.jpg)
21
Model-Based
DesignImplementing MBD Workflow
Model Management and
Architecture
Verification and Validation
Polyspace©
Polyspace Code ProverTM
STATEFLOW®
Event-Based Modeling
Code GenerationRapid Prototyping and HIL-
Simulation
Embedded Systems
FPGA Design
Generating HDL Code
Xilinx Zynq SoCs
AUTOSAR
MATLAB®
SIMULINK®
Application-
Specific Control System Design
Signal Processing
Communication Systems
LTE Systems
Image and Video
ProcessingImage Processing
Computer Vision
Computational
FinanceRisk Management
Time-Series Modelling
Application
DevelopmentProgramming Techniques
Building Interactive
Applications
Object-Oriented Programming
Data AnalyticsData Processing and
Visualization
Statistics
Machine Learning
Optimization Techniques
Parallel Computing
Code GenerationMATLAB Coder
Interfacing with C-code
Signal ProcessingUsing MATLAB
Using Simulink
Code IntegrationIntegrating C and MATLAB
SimscapeTM
General SimscapeTM
Simscape MultibodyTM
Simscape DrivelimeTM
Simscape FluidsTM
Simscape Power SystemsTM
https://nl.mathworks.com/services/training.html
![Page 22: Machine Learning Simplified€¦ · Consider Machine Learning When update as more data becomes available learn complex non-linear relationships learn efficiently from very large data](https://reader033.vdocuments.mx/reader033/viewer/2022050518/5fa1e8ee5a28783f4d66a15b/html5/thumbnails/22.jpg)
22© 2017 The MathWorks, Inc.
© 2017 The MathWorks, Inc. MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See www.mathworks.com/trademarks
for a list of additional trademarks. Other product or brand names may be trademarks or registered trademarks of their respective holders.