LATIDIA · Ciberseguridad

Hacia el DP certificado por TEE: capacitación privada diferencial verificable en GPU heredadas

arXiv:2609.20532v1 Tipo de anuncio: nuevo Resumen: La amplia adopción del aprendizaje automático ha creado una creciente demanda de políticas y regulaciones para proteger los datos confidenciales de capacitación, con la privacidad diferencial (DP) emergiendo como una clave

La noticia

arXiv:2609.20532v1 Announce Type: new Abstract: Wide adoption of machine learning has created growing policy and regulatory demand for protecting sensitive training data, with differential privacy (DP) emerging as a key mechanism. Yet a less-studied problem is how to certify the faithful execution of DP during training: an external verifier should be able to check that a released model was trained with proper DP protection, without accessing the private training data. Existing cryptographic approaches, such as zero-knowledge proofs, provide strong guarantees but often incur prohibitive overhead, in some cases by orders of magnitude. Trusted Execution Environments (TEEs) offer a more efficient alternative, but the multi-GPU TEE support needed for training and