The paper introduces ScoreAdv, a novel approach for generating natural adversarial examples (UAEs) using diffusion models. It incorporates an adversarial guidance mechanism and saliency maps to shift the sampling distribution and inject visual information. Experiments on ImageNet and CelebA datasets demonstrate state-of-the-art attack success rates, image quality, and robustness against defenses.
This paper introduces Diffusion-BBO, a new online black-box optimization (BBO) framework that uses a conditional diffusion model as an inverse surrogate model. The framework employs an Uncertainty-aware Exploration (UaE) acquisition function to propose scores in the objective space for conditional sampling. The approach is shown theoretically to achieve a near-optimal solution and empirically outperforms existing online BBO baselines across 6 scientific discovery tasks.
This paper introduces a self-supervised learning method for point cloud analysis using an upsampling autoencoder (UAE). The model uses subsampling and an encoder-decoder architecture to reconstruct the original point cloud, learning both semantic and geometric information. Experiments show the UAE outperforms existing methods in shape classification, part segmentation, and point cloud upsampling tasks.
This paper introduces a unified deep autoregressive model (UAE) for cardinality estimation that learns joint data distributions from both data and query workloads. It uses differentiable progressive sampling with the Gumbel-Softmax trick to incorporate supervised query information into the deep autoregressive model. Experiments show UAE achieves better accuracy and efficiency compared to state-of-the-art methods.
A KAUST-led team used acoustic technology and net sampling to explore ocean depths in 146 locations worldwide. They found that the biomass of pelagic organisms in the deep sea is much larger than previously thought, especially below productive waters. Deep-sea fauna migrates to shallower areas to feed, transporting carbon to the deep sea. Why it matters: This suggests the ocean's carbon pump is more effective than previously understood, with implications for climate change mitigation strategies.
A research paper by Dr. Reda Alami of the AI and Digital Science Research Center (AIDRC) at TII has been accepted for publication at the 14th Asian Conference on Machine Learning (ACML 2022). The paper addresses sequential decision-making under uncertainty in non-stationary environments, proposing a Bayesian Change-Point Detection with Thompson Sampling (Bayesian-CPD-TS) algorithm. The algorithm combines decision-making under uncertainty and sequential detection of abrupt changes. Why it matters: This recognition highlights the growing AI research capabilities within the UAE and its contribution to the global machine learning community.
Prof. Luigi Amico of the Quantum Research Center (QRC) has been awarded a US$250,000 Julian Schwinger Foundation (JSF) Grant for his 'Rydberg Atomtronics' project. The project involves experimental sampling managed by Prof. Oliver Morsch’s group in Pisa and will run in collaboration with the University of Catania and the University of Pisa. The grant will facilitate the development of quantum circuits using Rydberg excitations for fast quantum devices and quantum computing. Why it matters: This funding and collaboration strengthens the UAE's position in quantum computing research and fosters international partnerships in advanced technology.
The National Center for Vegetation Cover Development and Combating Desertification (NCVC) and KAUST have launched the SAUDINet initiative. The initiative aims to advance terrestrial ecology research in Saudi Arabia, focusing on restoring degraded lands, enhancing carbon sequestration and preserving biodiversity. NCVC’s workforce will receive specialized training in biodiversity monitoring and ecological sampling, with samples analyzed in KAUST’s labs. Why it matters: The partnership aims to establish Saudi Arabia as a global leader in the study of arid ecosystems and address the lack of data from hyper-arid lands in climate models.