We may earn an affiliate commission when you visit our partners.

Introduction to On-Device AI

As AI moves beyond the cloud, on-device inference is rapidly expanding to smartphones, IoT devices, robots, AR/VR headsets, and more. Billions of mobile and other edge devices are ready to run optimized AI models.

This course equips you with key skills to deploy AI on device:

1. Explore how deploying models on device reduces latency, enhances efficiency, and preserves privacy.

2. Go through key concepts of on-device deployment such as neural network graph capture, on-device compilation, and hardware acceleration.

This course equips you with key skills to deploy AI on device:

1. Explore how deploying models on device reduces latency, enhances efficiency, and preserves privacy.

2. Go through key concepts of on-device deployment such as neural network graph capture, on-device compilation, and hardware acceleration.

3. Convert pretrained models from PyTorch and TensorFlow for on-device compatibility.

4. Deploy a real-time image segmentation model on device with just a few lines of code.

5. Test your model performance and validate numerical accuracy when deploying to on-device environments

6. Quantize and make your model up to 4x faster and 4x smaller for higher on-device performance.

7. See a demonstration of the steps for integrating the model into a functioning Android app.

Learn from Krishna Sridhar, Senior Director of Engineering at Qualcomm, who has played a pivotal role in deploying over 1,000 models on devices and, with his team, has created the infrastructure used by over 100,000 applications.

By learning these techniques, you’ll be positioned to develop and deploy AI to billions of devices and optimize your complex models to run efficiently on the edge.

Enroll now

Or subscribe to Coursera Plus

And get unlimited access to Coursera

Two deals to help you save

Save money when you learn. Here are two deals and offers that may be relevant to this course.

All coupon codes, vouchers, and discounts are applied automatically unless otherwise noted.

Valid until December 2

Annual Subscription

Invest in your career growth and save. Start your annual subscription to Coursera Plus and access 10,000+ programs.

Valid until March 12, 2027

Save 40% Off 3 Months

Access thousands of top-rated learning programs with a Coursera Plus monthly subscription and take 40% off 3 months for a limited time.

What's inside

Syllabus

Traffic lights

Read about what's good

what should give you pause

and possible dealbreakers

Geared toward equipping learners with essential techniques to deploy machine learning models on devices

Taught by Krishna Sridhar, a recognized expert in on-device AI deployment with substantial experience in the field

Provides hands-on labs to reinforce learning and practical application of concepts

Aims to support learners in developing and deploying AI models efficiently on billions of devices

Covers key concepts in on-device AI deployment, including neural network graph capture and hardware acceleration

Suitable for learners interested in developing and deploying AI models specifically on devices

Save this course

Create your own learning path. Save this course to your list so you can find it easily later.

Save

Reviews summary

Practical introduction to on-device ai deployment

According to learners, this course offers a largely positive and practical introduction to on-device AI. Students commend the clear explanations by instructor Krishna Sridhar, whose industry experience at Qualcomm shines through, providing invaluable perspectives. The course is praised for its hands-on labs, especially the real-time image segmentation deployment and the crucial section on model optimization through quantization. Many found the content on PyTorch and TensorFlow model conversion and hardware acceleration highly beneficial. While it provides a solid foundation for deploying AI on edge devices, some reviewers noted that the Android app integration was more of a demo, and certain advanced topics could use more in-depth coverage for those seeking beyond an introduction. Overall, it's considered a must-take for professionals entering mobile or edge AI.

Explains fundamental concepts and benefits of on-device AI clearly.

"The content on PyTorch/TensorFlow model conversion and deployment was very helpful."

"The explanations of hardware acceleration and the benefits of edge deployment were clear."

"The benefits of edge deployment (latency, privacy) were well articulated."

Covers critical techniques like quantization for performance and size.

"The section on quantization was extremely valuable, showing how to optimize models for real-world deployment."

"I found the practical approach to model optimization (quantization) incredibly useful."

"The coverage of quantization and efficiency was crucial."

"I learned how to make my model up to 4x faster and 4x smaller for higher on-device performance."

Emphasizes hands-on model deployment and real-world application.

"I particularly appreciated the hands-on labs and the practical demonstration of integrating a model into an Android app."

"As a software engineer, I found the practical approach to model optimization and deployment incredibly useful."

"The segment on deploying a real-time image segmentation model was fantastic and very hands-on."

"It’s rare to find a course that bridges the gap between theory and actual deployment so effectively."

Benefits from the instructor's deep industry experience and clear explanations.

"The instructor, Krishna Sridhar, explains complex concepts with incredible clarity."

"The instructor's industry experience shines through."

"The instructor's expertise is evident. It demystified a lot of the complexities of deploying AI at the edge."

"The instructor's background from Qualcomm provided invaluable perspectives."

The Android integration is a demonstration, not an interactive lab.

"My only minor critique is that some parts felt a bit rushed, especially the Android integration which was more of a demo than a deep dive."

"The Android app demo was interesting but not interactive."

Provides a solid introduction but may lack depth for advanced learners.

"I came into this course with some prior AI knowledge, hoping for a deep dive into optimization... felt more like an 'introduction'."

"Good for a high-level understanding, but not enough for immediate practical application without further study."

"While it touched on quantization, it didn't go as deep as I wanted."

"This course might be a bit basic for intermediate users seeking advanced techniques."

Activities

Be better prepared before your course. Deepen your understanding during and after it. Supplement your coursework and achieve mastery of the topics covered in Introduction to On-Device AI with these activities:

Review TensorFlow Library

Show steps

Go through TensorFlow's documentation to refresh your understanding of the library, enabling you to effectively convert models.

Browse courses on Deep Learning

Show steps

Visit TensorFlow website and review their documentation
Practice writing TensorFlow code

Connect with On-Device AI Deployment Experts

Show steps

Guidance from experienced professionals can accelerate your learning and provide valuable insights into the field of on-device AI deployment.

Show steps

Identify potential mentors who have expertise in on-device AI deployment
Reach out to these individuals and express your interest in mentorship

Practice Neural Network Graph Capture Techniques

Show steps

Getting your hands dirty with these techniques will make you well-prepared for the core concepts of this course on on-device AI deployment.

Show steps

Read course materials about Neural Network Graph Capture
Implement a Neural Network Graph Capture function
Solve coding problems that require Neural Network Graph Capture

Ten other activities

Expand to see all activities and additional details

Show all 13 activities

Review models before course

Show steps

Review key machine learning theory and concepts (i.e. model training, validation, evaluation) before the start of the course.

Browse courses on Machine Learning

Show steps

Review key theoretical concepts: Model Training
Review key theoretical concepts: Model Validation
Review key theoretical concepts: Model Evaluation

Practice Neural Network Optimization

Show steps

Refresh your skills in neural network optimization to prepare for this course's focus on optimizing AI models for on-device deployment.

Show steps

Review your notes and assignments from previous machine learning or neural network optimization courses.
Follow tutorials on neural network optimization techniques.
Implement neural network optimization algorithms in a programming language of your choice.

Python C++ coding practice

Show steps

Complete practice problems using Python and C++ to strengthen coding skills necessary for deploying models on device.

Browse courses on Python

Show steps

Identify online coding practice problems
Attempt to solve coding problems
Review solutions and identify areas for improvement

Follow a Tutorial on Model Performance Testing and Validation

Show steps

Understanding how to test and validate models will increase your confidence while deploying AI on-device.

Show steps

Identify a guided tutorial that covers testing and validating models for on-device environments
Follow the tutorial and complete all exercises
Apply the techniques you learned to your own project

Explore On-Device AI Tutorials

Show steps

Supplement your learning by exploring tutorials and resources on on-device AI to gain a deeper understanding of the concepts and techniques covered in this course.

Show steps

Search for tutorials on topics such as on-device model deployment, optimization, and hardware acceleration.
Follow tutorials that align with your interests and skill level.
Experiment with the code and examples provided in the tutorials.

On-device inference with MobileNet

Show steps

Practice implementing on-device inference with MobileNet through online tutorials.

Browse courses on AI

Show steps

Identify tutorial on on-device inference with MobileNet
Follow the steps outlined in the tutorial
Test and evaluate the accuracy of your implementation

AI Model Deployment in Android

Show steps

Build an Android application that integrates an AI model for image recognition, putting your theoretical knowledge into practice.

Browse courses on Android Development

Show steps

Plan and design the Android application
Develop the Android application
Integrate the AI model into the application
Test and evaluate the application

Deploy a Real-Time Object Detection Model on a Smartphone

Show steps

Hands-on experience is one of the best ways to solidify the concepts learned in this course on on-device AI deployment.

Show steps

Gather the necessary hardware and software
Follow the course materials to deploy a real-time object detection model
Test and evaluate the deployed model

Technical Blog on Model Quantization

Show steps

Write a blog post explaining the process and effectiveness of quantizing models, to reinforce knowledge and showcase understanding.

Browse courses on Software Development

Show steps

Research model quantization
Write a technical blog describing your findings
Publish your blog and share it with others

Attend a Workshop on Quantization for On-Device AI Deployment

Show steps

Workshops offer an immersive learning experience that can greatly enhance your understanding of on-device AI deployment.

Show steps

Identify and register for a workshop on quantization for on-device AI deployment
Attend the workshop and participate actively
Apply the knowledge gained from the workshop to your own projects

Career center

Learners who complete Introduction to On-Device AI will develop knowledge and skills that may be useful to these careers:

Reading list

We've selected 50 books that we think will supplement your learning. Use these to develop background knowledge, enrich your coursework, and gain a deeper understanding of the topics covered in Introduction to On-Device AI.

Hands-On Machine Learning with Scikit-Learn, Keras,...

Save

Serves as the definitive modern hands-on standard for machine learning, guiding practitioners through concrete scikit-learn workflows before deeply exploring TensorFlow 2 and Keras architectures.

Introduction to On-Device AI

Two deals to help you save

What's inside

Syllabus

Traffic lights

Save this course

Reviews summary

Practical introduction to on-device ai deployment

Activities

Career center

Reading list

Share

Similar courses

Coursera Buyer Protection & Guarantee

7-Day Free Trial & Free Audit

14-Day Refund Guarantee

Verified Academic & Industry Partner