Hash Layers For Large Sparse Models

by   Stephen Roller, et al.

We investigate the training of sparse layers that use different parameters for different inputs based on hashing in large Transformer models. Specifically, we modify the feedforward layer to hash to different sets of weights depending on the current token, over all tokens in the sequence. We show that this procedure either outperforms or is competitive with learning-to-route mixture-of-expert methods such as Switch Transformers and BASE Layers, while requiring no routing parameters or extra terms in the objective function such as a load balancing loss, and no sophisticated assignment algorithm. We study the performance of different hashing techniques, hash sizes and input features, and show that balanced and random hashes focused on the most local features work best, compared to either learning clusters or using longer-range context. We show our approach works well both on large language modeling and dialogue tasks, and on downstream fine-tuning tasks.


page 1

page 2

page 3

page 4


Efficient Language Modeling with Sparse all-MLP

All-MLP architectures have attracted increasing interest as an alternati...

BASE Layers: Simplifying Training of Large, Sparse Models

We introduce a new balanced assignment of experts (BASE) layer for large...

Superbloom: Bloom filter meets Transformer

We extend the idea of word pieces in natural language models to machine ...

StableMoE: Stable Routing Strategy for Mixture of Experts

The Mixture-of-Experts (MoE) technique can scale up the model size of Tr...

DLB: Deep Learning Based Load Balancing

Load balancing mechanisms have been widely adopted by distributed platfo...

On Hash-Based Work Distribution Methods for Parallel Best-First Search

Parallel best-first search algorithms such as Hash Distributed A* (HDA*)...