Work

Synthesizing Antecedent and Defeater Structure in Exact Neural Causal Laws by Direct Editing and Training

Year: 2026
Type: preprint
Abstract: For a float32 ReLU classifier under a declared binary32 semantics, the classes it assigns to all interventions on a few pixels of a fixed image form a finite object, its finite intervention law; for one target class it is a Boolean function that can be enumerated exactly. We study how a prescribed structure of such a law can be synthesized. On a 784–128–10 MNIST classifier and one anchor image, the target is a predicate on the complete one-pixel and pixel-pair laws: no single-pixel change reaches class t before; afterwards exactly one designated pixel does, above a threshold, and a second pixel blocks it at high values. A closed-form rewrite of one hidden unit, inactive on all 70,000 MNIST images, introduces the predicate for all 9 non-anchor classes and changes no MNIST prediction. Ordinary training with one distinct crafted example, appended to 15–30 declared batches, introduces it for 8 classes, with the blocking pixel and schedule found by exhaustive search and a declared rule. Against a matched control that reproduces the baseline bit for bit, restoring the first-layer weights of either pixel shows that the training change is necessary for both the route and the block; a sham insertion indicates a target-specific response of the dataset for 6 of these classes. For classes 5 and 7, a one-unit edit fitted by a linear program and the trained model have identical laws of the target class on all one-pixel and pair interventions, although their parameter changes differ by orders of magnitude in size and support. All positive results are exhaustive binary32 checks of saved models.
Language: English
Open Access status: green
OpenAlex ID: W7225897260
Cites:
Cited by:
Related to: 10