IISc Bangalore · Computer Science

Computer System Optimizations for ML

Prof. Sumit K. Mandal · 60 free video lectures

Increasing complexity of machine learning (ML) algorithms have necessitated the emergence of specialized computer systems. Some ML algorithms are executed at edge, some are executed on the cloud. In this course, we will delve into different computing kernels of both training and inference of ML…

▶ Play on Open Lecture HallCourse page ↗

Lectures

  1. Introduction to the course
  2. Neural Networks
  3. Introduction to Convolutional Neural Network - 1
  4. Introduction to Convolutional Neural Network - 2
  5. Convolutional Neural Network-1
  6. Convolutional Neural Network-2
  7. Convolutional Neural Network-3
  8. Canonical CNN Architectures - 1
  9. Canonical CNN Architectures - 2
  10. CNN Computational Optimization Techniques
  11. Dataflow for CNN Execution - 1
  12. Dataflow for CNN Execution - 2
  13. CNN Accelerators - 1
  14. CNN Accelerators 2
  15. CNN Accelerators 3
  16. CNN Accelerators 4
  17. CNN prunning - 1
  18. CNN pruning - 2
  19. Parallelism in CNN - 1
  20. Parallelism in CNN - 2
  21. Parallelism in CNN - 3
  22. Parallelism in CNN - 4
  23. Parallelism in CNN - 5
  24. LLMs-1
  25. LLMs-2
  26. LLMs-3
  27. LLMs-4
  28. LLMs-5
  29. LLMs-6
  30. LLMs-7
  31. LLMs-8
  32. LLMs-9
  33. LLMs-10
  34. Distributed Service
  35. Fault Tolerance
  36. Fault-Tolerant Training
  37. Fundamentals of Graph Neural Networks-1
  38. Fundamentals of Graph Neural Networks-2
  39. Hardware Acceleration of Graph Neural Networks-1
  40. Hardware Acceleration of Graph Neural Networks -2
  41. ISSAC, Pipelayer
  42. Pipelayer[contd.], Atomlayer, Simulation platform
  43. Simulation platform, Area-aware optimization, communication-aware optimization
  44. Communication-Aware optimization, Achieving minimum communication latency
  45. Achieving minimum communication latency
  46. Advanced Hardware Architectures for GNN Acceleration
  47. Communication-Aware Pruning for IMC
  48. Re-Transformer
  49. Performance Evaluation of 2.5D IMC-Based AI Accelerators
  50. Chiplet-Based Heterogeneous Architectures for LLM Acceleration
  51. Chiplet-Based Heterogeneous Architectures for LLM Acceleration
  52. Performance Characterization
  53. Performance Characterization of Deep Learning Workloads - Part 1
  54. Performance Characterization of Deep Learning Workloads - Part 2
  55. Performance Characterization of Deep Learning Workloads - Part 3
  56. Introduction to AMD Neural Processing Unit (NPU)
  57. Introduction to AMD Neural Processing Unit (NPU)[contd.]
  58. Acceleration of ML operation with GPUs
  59. Training Large model with limited GPU memory
  60. Recap of the course

From NPTEL (IITs and IISc) · CC BY-NC-SA. More from IISc Bangalore.

Open Lecture Hall is free, with no ads and no sign-up. It is not run by or connected to any university; the lectures are the universities' own free recordings, played from where they host them. About & rights