Newest papers

Interflow: Breaking the Offline-Online Barrier for Private AI at the Edge

Hanieh Totonchi Asl, Ebrahim Nouri, Zhuoran Li, Danella Zhao · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.27431 · cited by 0 works (OpenAlex) · open-access PDF

French

L’interconnexion des données de l’apprentissage mécanique (PPML) à l’intermédiaire de l’interconnexion des données de l’apprentissage mécanique à l’intermédiaire de l’interconnexion des données de l’intermédiaire de l’interconnexion (PPML) offre une voie prometteuse vers l’apport des services d’IA puissants à l’extrémité sans exposer les données d’utilisateur sensibles. Cependant, son déploiement pratique reste restreint par les coûts élevés de communication, de calcul et de mouvement de données de l’interconnexion sécurisée multiparty computing (MPC), en particulier lors d’une inference non linéaire. Les systèmes

Machine translation (Meta M2M100); the English original follows.

Abstract

Privacy-preserving machine learning (PPML) offers a promising path toward bringing powerful AI services to the edge without exposing sensitive user data. However, its practical deployment remains constrained by the high communication, computation, and data-movement costs of secure multi-party computation (MPC), particularly during nonlinear inference. Existing MPC systems typically decouple input-independent correlated randomness generation from online execution, creating substantial memory pressure to buffer single-use cryptographic data. Moreover, limited storage can force a stop-and-refill execution pattern that repeatedly stalls the pipeline to replenish these buffers, bringing randomness generation back onto the critical path. We propose Interflow, an algorithm-hardware co-design that rethinks this separation by batch-wise processing and pipelining correlated-randomness generation with online execution: while the current batch consumes pre-generated randomness to execute secure nonlinear operations, correlation generation proceeds concurrently for the next batch. By leveraging interleaved execution, efficient data movement, and hardware acceleration, the design transforms preprocessing from a serialized, storage-intensive offline phase into a continuously overlapped pipeline that generates correlated randomness closer to the point of consumption, reducing memory footprint and data movement overhead. Implemented on the Zynq UltraScale+ ZCU102, Interflow improves nonlinear evaluation throughput by up to 1.55x for CNN operators and 1.48x for transformer operators. It achieves up to 1.26x speedup for ResNet-50 and up to 1.21x speedup for BERT-base, and reduces memory footprint by up to 37%, compared with baseline frameworks.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Speedup, Randomness, Pipeline (software), Memory footprint, Throughput

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange