×
DeepMind’s “PEER” Architecture Scales Language Models While Keeping Costs Low
Written by
Published on
Join our daily newsletter for breaking news, product launches and deals, research breakdowns, and other industry-leading AI coverage
Join Now

DeepMind’s new Mixture-of-Experts (MoE) architecture, PEER, scales language models to millions of tiny experts, improving performance while keeping computational costs down.

Key innovation: Parameter Efficient Expert Retrieval (PEER); DeepMind’s novel MoE architecture introduces a learned index to efficiently route input data to a vast pool of millions of tiny experts, enabling significant scaling without slowing down inference:

  • PEER replaces the fixed router in traditional MoE with a fast initial computation to create a shortlist of potential expert candidates before activating the top experts.
  • Unlike previous MoE architectures with large experts, PEER uses tiny single-neuron experts in the hidden layer, allowing knowledge sharing among experts and improving parameter efficiency.
  • PEER employs a multi-head retrieval approach, similar to the multi-head attention in transformer models, to compensate for the small expert size.

Overcoming MoE scaling limitations: Current MoE techniques are limited to a relatively small number of experts due to fixed routers that need readjustment when new experts are added:

  • Studies suggest increasing MoE “granularity” (number of experts) can improve performance, especially with increased model size and training data.
  • High-granularity MoE also enables models to learn new knowledge more efficiently by adding experts and proper regularization.
  • PEER addresses these challenges, allowing MoE to scale to millions of experts without the limitations of fixed routers.

Outperforming dense models and other MoEs: Experiments show PEER models achieve a better performance-compute tradeoff compared to transformer models with dense feedforward layers and other MoE architectures:

  • PEER reaches lower perplexity scores with the same computational budget as counterparts.
  • Increasing the number of experts in PEER leads to further perplexity reduction, challenging the belief that MoE efficiency peaks with a limited number of experts.

Broader implications for large language models: PEER’s approach can help further reduce the cost and complexity of training and serving very large language models:

  • PEER might be used in DeepMind’s Gemini 1.5 models, which reportedly use “a new Mixture-of-Experts (MoE) architecture.”
  • The architecture shows potential for dynamically adding new knowledge and features to LLMs by adapting PEER to select parameter-efficient fine-tuning adapters at runtime.
  • As the race to scale LLMs continues, PEER represents a significant step in improving the performance-compute tradeoff, positioning it as a competitive alternative to dense layers in foundation models.
DeepMind’s PEER scales language models with millions of tiny experts

Recent News

Could automated journalism replace human journalism?

This experimental AI news site combines automation with journalistic principles, producing over 20 daily articles at just 30 cents each while maintaining factual accuracy and source credibility.

Biosecurity concerns mount as AI outperforms virus experts

AI systems demonstrate superior practical problem-solving in virology laboratories, posing a concerning dual-use scenario where the same capabilities accelerating medical breakthroughs could provide step-by-step guidance for harmful applications to those without scientific expertise.

How AI is transforming smartphone communication

AI capabilities are now being embedded directly into existing messaging platforms, eliminating the need for separate apps while maintaining conversational context for more efficient communication.