Learn · NPTEL

Course10610623851 lectures

Applied Accelerated Artificial Intelligence

Prof. Satyajit Das and Prof. Satyadhyan Chickerur · IIT Guwahati KLE Technological University

Start with lecture 1 At NPTEL

About

Technical Partners

ABOUT THE COURSE: This course will cover the fundamentals of the compute capabilities and the system software required for implementing artificial intelligence (AI) based solutions with accelerated computing on industrial use cases such as the one in the domains of Smart City. The course will discuss end to end deployments of industrial use cases like using large language models with demonstration, and hence will help participants use state-of-the-art AI SDKs effectively to solve complex problems.

The content was originally developed for a course organized by the NSM Nodal Centers at IIT Goa, IIT Kharagpur, IIT Madras, and IIT Palakkad. (Link: https://iitgoa.ac.in/aishikshaai/)

Intended audience: Ph.D Scholars (any stream of Science or Engineering); Post Graduate Students (any stream of Science or Engineering); 3rd and 4th year UG Students (any stream of Engineering); Non-Students and Working Professionals

Prerequisites: Prior knowledge of Computer Organization, High-Performance; Computing, Machine Learning and Deep learning is desirable

Lectures

Week 1

  1. Lecture 1: CPU Architecture YouTube
  2. Lecture 2 : Memory Hierarchy YouTube
  3. Lecture 3 : GPU Fundamentals & Modern YouTube
  4. Lecture 4: Interconnects - PCIe, NVLink and NVSwitch YouTube
  5. Lecture 5 : AI Software stack - OS, Drivers and Runtime Layers YouTube

Week 2

  1. Lecture 1 : Docker fundamentals for GPU - enabled AI workloads YouTube
  2. Lecture 2 : Deterministic Environment & Containerized Jupyter YouTube
  3. Lecture 3 : CUDA in Containers & CI Validation YouTube
  4. Lecture 4 : Schedulers & GPU Allocation in Multi-User Environments YouTube

Week 3

  1. Lecture 1 : Kubernetes Fundamentals for AI Workloads YouTube
  2. Lecture 2 : GPU Pod Scheduling,Node Selector, Tolerations YouTube
  3. Lectrue 3 : K8s Networking & Configuration for AI Workloads YouTube
  4. Lecture 4 : ConfigMaps, Secrets & Config Injection YouTube

Week 4

  1. Lecture 1 : Tensors & Autograd YouTube
  2. Lecture 2 : Building Models with nn.Module YouTube
  3. Lecture 3 : The Training Loop YouTube
  4. Lecture 4 : Data loaded & Feeding the model efficiently YouTube
  5. Lecture 5 : First steps in performance YouTube

Week 5

  1. Lecture 1 : TensorFlow Execution Models & Performance Basics YouTube
  2. Lecture 2 : Optimization Input Pipelines with tf.data YouTube
  3. Lecture 3 : Graph Execution & TensorFlow Optimization YouTube
  4. Lecture 4 : Introduction to XLA Compilation YouTube
  5. Lecture 5 : Model Interoperability: SavedModel vs ONNX YouTube

Week 6

  1. Lecture 1 : LLM Foundation and terminology YouTube
  2. Lecture 2 : Mixed precision and numerical stability YouTube
  3. Lecture 3 : Memory Optimization and throughput YouTube
  4. Lecture 4 : Transfer learning for large models YouTube
  5. lecture 5 : Efficient LLM fine-tuning using PEFT AND LoRA YouTube

Week 7

  1. Fundamentals of Distributed AI Computing Session 1 Part 1 YouTube
  2. Fundamentals of Distributed AI Computing Session 1 Part 2 YouTube
  3. Fundamentals of Distributed AI Computing Session 2 Part 1 YouTube
  4. Fundamentals of Distributed AI Computing Session 2 Part 2 YouTube
  5. Distributed Deep Learning using Tensorflow and Horovod YouTube

Week 8

  1. Challenges with Distributed Deep Learning Training Convergence YouTube
  2. Fundamentals of Accelerating Deployment part 1 YouTube
  3. Fundamentals of Accelerating Deployment part 2 YouTube

Week 9

  1. Accelerating neural network inference in PyTorch and TensorFlow part 1 YouTube
  2. Accelerating neural network inference in PyTorch and TensorFlow part 2 YouTube
  3. Accelerated Data Analytics part 1 YouTube
  4. Accelerated Data Analytics part 2 YouTube
  5. Accelerated Data Analytics part 3 YouTube
  6. Accelerated Data Analytics part 4 YouTube
  7. Accelerated Machine Learning YouTube

Week 10

  1. Introduction to NLP Part 1 YouTube
  2. Introduction to NLP Part 2 YouTube

Week 11

  1. Applied AI: Smart City ( Intelligent Video Analytics) Session 1 part 1 YouTube
  2. Applied AI: Smart City ( Intelligent Video Analytics) Session 1 part 2 YouTube
  3. Applied AI: Smart City ( Intelligent Video Analytics) Session 2 Deepstream part 1 YouTube
  4. Applied AI: Smart City ( Intelligent Video Analytics) Session 2 Deepstream part 2 YouTube

Week 12

  1. Introduction to word embedding YouTube
  2. Text classification using word embedding YouTube

Syllabus and materials

At NPTEL, as the course publishes them.

Credits and licence

Institution
IIT Guwahati KLE Technological University
Department
Computer Science and Engineering
Taught by
Prof. Satyajit Das, Prof. Satyadhyan Chickerur
Level
Graduate
Language
English
Subjects
Computer science, Engineering
Source
NPTEL
Licence
CC BY-NC-SA, as NPTEL states it. Non-commercial use only. NPTEL's terms
Attribution
NPTEL states: "Distributed under Creative Commons Attribution-ShareAlike / CC BY - NC - SA", beside a BY-NC-SA badge. Coordinated by IIT Guwahati.
Who can watch
NPTEL gives no rating. Invisible Network shows it to Tweens profiles and older, judged from what the source says about it. How courses are placed