Mixture of Experts: Core Concepts and Hands-on Implementation
Chapter 1: Foundations of Mixture of Experts Models
Overview of Sparsely-Gated MoE Architecture
The Gating Network: Formulation and Function
Expert Networks: Specialization and Capacity
Mathematical Formulation of the MoE Layer
Load Balancing and Auxiliary Losses
Challenges in MoE Training: Expert Collapse
Comparison with Dense Model Scaling
Hands-on: Implementing a Basic MoE Layer
Chapter 2: Advanced Routing Mechanisms
Analysis of Top-k Gating and its Variants
Noisy Top-k Gating for Load Balancing
Hash-based Routing for Deterministic Selection
Switch Transformers: Simplified Routing
Soft MoE: Differentiable Routing
Analyzing Routing Decisions and Specialization
Hands-on: Implementing Different Routing Strategies
Chapter 3: Training and Optimization of Large-Scale MoEs
Expert Parallelism for Distributed Training
Combining Model, Data, and Expert Parallelism
Capacity Factor and its Impact on Performance
Techniques for Mitigating Router Z-Loss Instability
Precision and its Effects: BFloat16 Training
Fine-tuning Strategies for Pre-trained MoE Models
Practice: Configuring a Distributed Training Job
Chapter 4: Efficient Inference with MoE Models
Inference Challenges: Memory and Latency
Expert Offloading to CPU or NVMe
Batching Strategies for Sparse Activation
Model Distillation for MoE Compression
Quantization Techniques for MoE Layers
Speculative Decoding with MoE Models
Hands-on: Building an Optimized Inference Pipeline
Chapter 5: Integrating MoE into Modern Architectures
Replacing FFNs with MoE Layers in Transformers
Placement of MoE Layers: Frequency and Location
MoE in Vision Transformers (ViT)
MoE in Multi-modal Models
Architectural Variants and their Properties
Analyzing Parameter vs. FLOPs Trade-offs
Practice: Modifying a Transformer to use MoE
Switch Transformers: Simplified Routing
Was this section helpful?
- Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity, William Fedus, Barret Zoph, Noam Shazeer, 2022 Journal of Machine Learning Research, Vol. 23 (JMLR, Inc.) - Introduces the Switch Transformer architecture with its simplified top-1 routing for efficient scaling.
- Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer, Noam Shazeer, Azalia Mirhoseini, Krzysztof Maziarz, Andy Davis, Quoc Le, Geoffrey Hinton, Jeff Dean, 2017 International Conference on Learning Representations (ICLR) DOI: 10.48550/arXiv.1701.06538 - Foundational paper on Mixture-of-Experts, outlining the general concept of sparsely activated models.
- MegaBlocks: Efficient Sparse Training with Mixture-of-Experts, Shaden Smith, Brandon Norick, Sam Ade Jacobs, Jonathan Frankle, Jeremy Gray, Elias Frantar, Tal Ben-Nun, Dan Alistarh, 2023 International Conference on Machine Learning (ICML), Vol. 202 - Focuses on memory-efficient and computationally optimized training for large sparse MoE models.