Science

A new approach to neural network design, leveraging Boolean logic instead of continuous values, dramatically improves resilience to bit-flip errors in parameters.

New research explores how combining robustness and uncertainty quantification offers a more comprehensive approach to evaluating the dependability of machine learning models.

A new protocol enables secure and efficient outsourcing of control computations using cryptographic techniques.

Modern software relies heavily on cryptography, and keeping those systems secure requires a proactive, architecture-driven approach to managing and migrating cryptographic components.

New research reveals that fully homomorphic encryption, despite its promise of secure computation, is surprisingly susceptible to subtle data corruption arising from hardware faults.

A new approach combines quantum key distribution with model predictive control to enhance privacy and reduce computational demands in networked control applications.

A new framework leverages the principles of quantum optimization to enhance the speed and reliability of microservice deployments across cloud and edge environments.

As artificial intelligence systems increasingly rely on learned memories, a new vulnerability emerges: the deliberate corruption of those memories, with potentially devastating consequences.

A new runtime system, Accompanist, leverages decentralized sagas to improve the performance and scalability of microservice architectures.

A new approach to object bounds protection, PRISM, offers a compelling balance between performance and precision in safeguarding against memory errors.