unearth.wiki

Machine Unlearning

/məˈʃiːn ʌnˈlɜːnɪŋ/ Coined in 2015 by Yinzhi Cao and Junfeng Yang in their foundational IEEE S&P paper. The computational discipline of selective forgetting.
Definition The computational subdiscipline and suite of algorithmic techniques aimed at selectively removing the influence of specific training data points from a trained model without retraining the entire architecture from scratch. In production foundation models, exact unlearning is computationally ruinous while approximate unlearning confronts an intractable verification limit.

The Problem of Selective Forgetting

Training an artificial neural network is a non-linear, irreversible optimization process. Gradient descent adjusts billions of floating-point parameters simultaneously based on the full training corpus. Information does not reside in a localized address or modular slot; it is distributed across the weight matrix. When personal or sovereign data is ingested, it is metabolized into the model's global representations.

Standard data protection laws (such as GDPR Article 17) grant data subjects a right to erasure without undue delay (typically interpreted as 30 days). For frontier models costing hundreds of millions of dollars and months of continuous compute to train, retraining from scratch per deletion request is economically impossible. Machine unlearning was conceived to bridge this gap.

Architectures of Deletion

Over a decade of research, computer science developed three primary approaches to unlearning:

The Verification Wall

Approximate unlearning encounters a fundamental epistemological limit: a model that unlearned a data point and a model that never saw that data point can produce identical parameter states. A party claiming to have executed machine unlearning can present weights without having performed any real removal, because forgetting is unfalsifiable from the model artifact itself.

Exact erasure is possible in principle and ruinous in practice, while approximate erasure is practical in every respect except the one that matters: no data subject, auditor, or court can verify that it occurred.

Intersection with the Baked-In Paradox

Because reliable, verifiable unlearning at foundation scale does not exist, claims of compliance default to the Baked-In Paradox: once semantic capital is absorbed into neural parameters, the contamination is structural and permanent. When companies use input/output filters to mimic unlearning without altering weights, the practice constitutes Erasure Theater, leaving Weight Incarceration active.

Field Notes & Ephemera

Field Note: "The operator claimed the biographical record had been 'unlearned' via weight projection. When we probed the latent representations with orthogonal steering vectors, the model completed the private phone number with 98% token confidence. The parameter had not forgotten; it had merely been silenced."
Stratigraphy (Related Concepts)
Baked-In Paradox Erasure Theater Architectural Repeal Algorithmic Disgorgement Weight Incarceration Ontology of the File Evidentiary Obfuscation

a liminal mind meld collaboration

unearth.im | archaeobytology.org