Deterministic equivalent and error universality of deep random features learning

Open in new window