Multi-head attention

As explained in Deep Learning with R, Third Edition

MultiHeadAttention is a Keras layer that computes attention through multiple parallel attention heads and combines their outputs.

Multi-head attention runs attention several times in parallel with different learned projections, then concatenates the head outputs. Separate heads can focus on different relationships, such as sentence subjects or punctuation, instead of washing all source information into one softmax sum. section 15.3.1

A reusable Keras layer

A multi-head attention layer can replace the explicit per-head implementation: multi_head_attention <- layer_multi_head_attention(num_heads = num_heads, head_dim = head_dim); multi_head_attention(query = target, key = source, value = source). Each head projects queries, keys, and values, computes scaled dot-product attention, and contributes a partition of the final output. section 15.3.1

With self-attention, query, key, and value all come from the same sequence, allowing every token to incorporate contextual information from every other token. Transformer encoder blocks combine this layer with residual connections, layer normalization, and a two-layer feed-forward network. section 15.3.2 section 15.3.2

Figure 15.7 Multiheaded attention allows each target word to attend to different parts of the source sequence in separate partitions of the eventual output vector.

Shows how separate attention heads let each target token attend to different source regions within separate output partitions. section 15.3.1

How Deep Learning with R, Third Edition covers it

How deep it goes

Explains projections, scaled scores, softmax weighting, concatenation, and encoder integration, without deriving the underlying mathematics.

Examples

Uses R/Keras listings, source-target sequence attention, and the contextual example of “station” attending to “train.”

What it assumes

Assumes sequence representations and attention basics while motivating the transition from RNN-based translation models to self-attention.

What you can do after

Build it from scratch or operate a Keras layer for cross-sequence attention, self-attention, and Transformer encoder blocks.

How it connects

Connects multi-head attention to dot-product attention, self-attention, RNNs, layer normalization, and Transformer encoder blocks.

liveBookmanning.com

Read Deep Learning with R, Third Edition on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Natural Language Processing in Action, Second EditionMulti-head attention computes multiple attention representations in parallel and combines them, potentially using distinct query and source sequences.
• 

  liveBookmanning.com
  Build a Text-to-Image Generator (from Scratch)An attention mechanism that computes attention through multiple separate query, key, and value subspaces to capture different relationships simultaneously.
• 

  liveBookmanning.com
  Transformers in ActionMultihead attention is an attention mechanism that computes multiple attention patterns in parallel so a model can represent different relationships among sequence positions.
• 

  liveBookmanning.com
  Learn Generative AI with PyTorchA self-attention variant that splits representations into multiple attention heads to capture different relationships in parallel.
• 

  liveBookmanning.com
  Inside Deep Learning: Math, Algorithms, ModelsAn attention mechanism that uses multiple attention heads to attend to different contexts or aspects of the input in parallel.
• 

  liveBookmanning.com
  Deep Learning with Python, Third EditionAn attention variant that runs several attention computations in parallel with different parameters to learn different relationships.
• 

  liveBookmanning.com
  Graph Neural Networks in ActionAn attention technique that computes multiple attention operations in parallel and combines their outputs.
• 

  liveBookmanning.com
  Build a Large Language Model (From Scratch)An attention technique that computes several independent attention heads in parallel and combines their outputs to capture different relationships in the same sequence.
• 

  liveBookmanning.com
  Time Series Forecasting Using Foundation ModelsAn attention mechanism with multiple parallel attention heads that combines information from different representation subspaces.
• 

  liveBookmanning.com
  Machine Learning Algorithms in DepthMultihead attention runs multiple attention mechanisms in parallel and combines their output representations.
• 

  liveBookmanning.com
  Deep Learning with PyTorch, Second EditionAn attention mechanism that runs several attention heads in parallel, each learning different projections and relationships.
• 

  liveBookmanning.com
  Optimization Algorithms: AI techniques for design, planning, and control problemsMulti-head self-attention computes several attention operations in parallel so a sequence representation can capture different relationships among its elements.
• 

  liveBookmanning.com
  The AI Pocket BookA transformer attention mechanism that splits representations into multiple heads to attend to different relationships in parallel.

Compare how 14 books explain multi-head attention →

Related concepts

• graph attention networkused by · A graph attention network is a graph neural network that uses attention weights to aggregate information from neighboring nodes, allowing their contributions to differ.
• grouped-query attentionequivalent to · Grouped-query attention partitions query heads into groups whose members share key and value projections, reducing attention computation and memory use.
• attentionkind of · Attention is a neural-network mechanism that computes input-dependent weights to combine information according to its relevance to a computation.
• attention headhas part · An attention head is an independently parameterized attention computation that evaluates relationships among input elements and combines information from them into a contextualized representation.
• transformer architectureused by · A transformer architecture is a neural-network architecture that processes sequences or other structured inputs using attention mechanisms and related transformations.
• encoderpart of · An encoder is a component that transforms input data into an encoded representation for subsequent processing, reconstruction, comparison, or transmission.

This explainer was generated by Manning's knowledge graph from the text of Deep Learning with R, Third Edition. The section links point to the parts of the book it draws on.