Activities - Compute Node Representations

#	Name	Abbrev.	Guidance	Actions
Data Preparation 3
1	Assemble initial node state	CRS-1	Concatenate the node name embedding with the type representation. Pad the result with zeros to …
1	Compute subtoken embeddings	CRS-1	Split each node name on camelCase and pascal_case boundaries into subtokens (e.g. classTypes → [class, …	Read the 'CRS-1 Compute subtoken embeddings' activity from playbook 'Program Graphs for Code Analysis' (workflow 'Compute Node Representations'). Follow its guidance when assisting me. Split each node name on camelCase and pascal_case boundaries into subtokens (e.g. classTypes → [class, types]). Learn an embedding for each subtoken. Compute the node name embedding as the average of its subtoken embeddings.
1	Compute type hierarchy embeddings	CRS-1	For each variable, collect the full set of its supertypes (type lattice τ(v)). Learn an …	Read the 'CRS-1 Compute type hierarchy embeddings' activity from playbook 'Program Graphs for Code Analysis' (workflow 'Compute Node Representations'). Follow its guidance when assisting me. For each variable, collect the full set of its supertypes (type lattice τ(v)). Learn an embedding r(τ) for each known type; use "UNKTYPE" for unseen types. Compute the variable's type representation as the element-wise maximum over {r(τ) : τ ∈ τ(v)}. During training, randomly sample a non-empty subset of τ*(v) as regularization.

Activities in Compute Node Representations