Advanced JAX: Performance, Optimization, and Scale
Chapter 1: Advanced JAX Transformations and Control Flow
Review of Core Transformations: jit, grad, vmap
Mastering lax.scan for Sequential Operations
Conditional Execution with lax.cond
Looping with lax.while_loop
Combining Control Flow and Transformations
Advanced Masking Techniques
Understanding Closures and JAX Staging
Practical: Implementing Complex Recurrent Logic
Chapter 2: Optimizing JAX Code for Performance
Profiling JAX Code on CPU, GPU, and TPU
Understanding JAX Computation Graphs (jaxpr)
The Role of XLA Compilation
Memory Layout and Its Impact on Performance
Avoiding Recompilation
Fusion and Operator Optimization
Asynchronous Dispatch
Practice: Optimizing a Numerical Computation
Chapter 3: Distributed Computing with JAX
Introduction to Parallelism Concepts
Device Management in JAX
Single-Program Multiple-Data (SPMD) with pmap
Implementing Data Parallelism using pmap
Collective Communication Primitives (psum, pmean, etc.)
Handling Axis Names in pmap
Nested pmap and Advanced Partitioning
Introduction to Multi-Host Programming
Practice: Distributed Data-Parallel Training
Chapter 4: Advanced Automatic Differentiation Techniques
Review of Forward- and Reverse-Mode Autodiff
Jacobian-Vector Products (JVPs) with jax.jvp
Vector-Jacobian Products (VJPs) with jax.vjp
Higher-Order Derivatives
Computing Full Jacobians and Hessians
Custom Differentiation Rules with jax.custom_vjp
Custom Differentiation Rules with jax.custom_jvp
Differentiation through Control Flow Primitives
Handling Non-Differentiable Functions
Practice: Implementing a Custom Gradient
Chapter 5: Interoperability and Custom Operations
Integrating JAX with NumPy
Zero-Copy Data Sharing with DLPack
Calling External CPU/GPU Code with jax.experimental.host_callback
Using jax.pure_callback for Side-Effect Free Calls
Introduction to JAX Primitives
Defining Custom Primitives
Implementing Abstract Evaluation Rules
Implementing Lowering Rules for Backends (CPU/GPU/TPU)
Defining Differentiation Rules for Custom Primitives
Practice: Integrating a C++ Function
Chapter 6: Large-Scale Model Training Techniques
Overview of Challenges in Large Model Training
Introduction to JAX Ecosystem Libraries (Flax, Haiku)
Managing Model Parameters and State
Combining pmap with Training Frameworks
Gradient Accumulation
Gradient Checkpointing (Re-materialization)
Mixed Precision Training
Model Parallelism Strategies
Optimization Algorithms for Large Scale
Practice: Implementing Gradient Checkpointing
Optimization Algorithms for Large Scale
Was this section helpful?
- Decoupled Weight Decay Regularization, Ilya Loshchilov, Frank Hutter, 2019 International Conference on Learning Representations (ICLR) DOI: 10.48550/arXiv.1711.05101 - Introduces AdamW, demonstrating that decoupling weight decay from adaptive gradient updates improves generalization for adaptive optimizers.
- Large Batch Optimization for Deep Learning, Yang You, Jing Li, Sashank Reddi, Jonathan Hseu, Sanjiv Kumar, Srinadh Bhojanapalli, Xiaodan Song, James Demmel, Kurt Keutzer, Cho-Jui Hsieh, 2020 International Conference on Learning Representations (ICLR) DOI: 10.48550/arXiv.1904.00962 - Proposes LAMB, an optimizer designed for stable and efficient training of deep learning models with extremely large batch sizes.
- Optax: A JAX library for gradient processing and optimization, DeepMind, 2024 - Official documentation for Optax, providing practical guidance and API reference for using various optimization algorithms and schedules within JAX.
- Attention Is All You Need, Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, Illia Polosukhin, 2017 Advances in Neural Information Processing Systems (NeurIPS), Vol. 30 DOI: 10.48550/arXiv.1706.03762 - Introduces the Transformer architecture and its associated learning rate schedule (warm-up followed by inverse square root decay), widely adopted for large language models.