MBZUAI researchers presented a new machine learning method at ICLR for uncovering hidden variables from observed data. The method, called "complementary gains," combines two weak assumptions to provide identifiability guarantees. This approach aims to recover true latent variables reflecting real-world processes, while solving problems efficiently. Why it matters: The research advances disentangled representation learning by finding minimal assumptions necessary for identifiability, improving the applicability of AI models to real-world data.
Researchers from MBZUAI introduce Forget-MI, a machine unlearning method tailored for multimodal medical data, enhancing privacy by removing specific patient data from AI models. Forget-MI utilizes loss functions and perturbation techniques to unlearn both unimodal and joint data representations. The method demonstrates superior performance in reducing Membership Inference Attacks and improving data removal compared to existing techniques, while preserving overall model performance and enabling data forgetting.
MBZUAI researchers have developed a "divide-and-conquer" technique to improve learning from demonstration in robotics. The approach breaks down complex dynamical systems into independently solvable subsystems, modeled as linear parameter-varying systems. This method aims to simplify computations while maintaining stability and accurately capturing joint interactions for robots in complex environments. Why it matters: The research addresses a key challenge in robotics, potentially enabling more efficient and safer robot learning from human demonstrations.
Professor Christian Roos from the University of Innsbruck will present a seminar on investigating many-body physics using trapped ions dressed by laser light. The talk will cover encoding, manipulating, and entangling quantum information using laser-cooled trapped ions. Recent experiments demonstrating entangling interactions and single-qubit control in strings of up to 50 ions will be discussed. Why it matters: This seminar contributes to advancements in quantum computing and simulation, fields of increasing interest in the Middle East as the region invests in advanced technologies.
Researchers at the Quantum Research Center's (QRC) Quantum Communications lab (QComms) achieved a milestone by demonstrating the violation of a Bell inequality using pairs of polarized entangled photons. This achievement serves as a "quantum health check" for their entangled photon source. The QRC team is working to harness entanglement effects in ultra-secure key distribution schemes to enhance secure communication. Why it matters: This advancement validates the application-readiness of QRC's quantum communication devices, paving the way for enhanced security in communication technologies within the region.