A single-channel 10b pipelined SAR ADC with a gm-cell residue amplifier and a current-mode fine SAR ADC achieves a 500MS/s conversion rate in a 28nm CMOS process under a 1.0 V supply. With background offset and gain calibration, the prototype ADC achieves an SNDR of 56.6dB at Nyquist. With power consumption of 6mW, it obtains a FoM of 21.7fJ/conversion-step.
We study the association between physical appearance and family income using a novel data which has 3-dimensional body scans to mitigate the issue of reporting errors and measurement errors observed in most previous studies. We apply machine learning to obtain intrinsic features consisting of human body and take into account a possible issue of endogenous body shapes. The estimation results show that there is a significant relationship between physical appearance and family income and the associations are different across the gender. This supports the hypothesis on the physical attractiveness premium and its heterogeneity across the gender.
Modeling a 3D face model that resembles a human face has been an important issue in computer graphics. A number of researchers have proposed various modeling methodologies to model the human face realistically. However, despite the fact that these methods give high-quality results, the time, cost, and labor required to generate such models by these methods are often high. In this paper, we propose a fully automatic face modeling method that can rapidly create a realistic 3D face model on the basis of a single 2D image. We studied 200 frontal facial images and captured their statistical shape variations. Using these properties, facial features were detected from the images. The depth information of the facial features was then estimated using an anthropometrical measurement database of human faces. Finally, by aligning the feature points three-dimensionally and interpolating them, a facial surface is generated. Finally, upon these 3-dimensional alignment of the facial features, a surface interpolates these feature points are created. As a result, we could obtain realistic 3D faces models of several individuals without any efforts. Further, the computational cost of our method was much lower than that of other methods. Thus, our method showed remarkably fast performance.
This article provides a brief overview of 18 projects funded in Track D—Data and Model Sharing to Enable AI Innovation—of the 2020 Cohort of the National Science Foundation's (NSF) Convergence Accelerator (CA) program. The NSF CA is focused on transitioning research to practice for societal impact. The projects described here were funded for one year in phase I of the program, beginning September 2020. Their focus is on delivering tools, technologies, and techniques to assist in sharing data as well as data-driven models to enable AI innovation. A broad range of domain areas is covered by the funded efforts, spanning across healthcare and medicine, to climate change and disaster, and civil/built infrastructure. The projects are addressing sharing of open as well as sensitive/private data. In September 2021, six of the eighteen projects described here were selected for phase II of the program, as noted in this article.
Federated learning (FL) enables multiple clients with distributed data sources to collaboratively train a shared model without compromising data privacy. However, existing FL paradigms face challenges due to heterogeneity in client data distributions and system capabilities. Personalized federated learning (pFL) has been proposed to mitigate these problems, but often requires a shared model architecture and a central entity for parameter aggregation, resulting in scalability and communication issues. More recently, model-heterogeneous FL has gained attention due to its ability to support diverse client models, but existing methods are limited by their dependence on a centralized framework, synchronized training, and publicly available datasets. To address these limitations, we introduce Federated Peer-Adaptive Ensemble Learning (FedPAE), a fully decentralized pFL algorithm that supports model heterogeneity and asynchronous learning. Our approach utilizes a peer-to-peer model sharing mechanism and ensemble selection to achieve a more refined balance between local and global information. Experimental results show that FedPAE outperforms existing state-of-the-art pFL algorithms, effectively managing diverse client capabilities and demonstrating robustness against statistical heterogeneity.
Abstract This article provides a brief overview of 18 projects funded in Track D—Data and Model Sharing to Enable AI Innovation—of the 2020 Cohort of the National Science Foundation's (NSF) Convergence Accelerator (CA) program. The NSF CA is focused on transitioning research to practice for societal impact. The projects described here were funded for one year in phase I of the program, beginning September 2020. Their focus is on delivering tools, technologies, and techniques to assist in sharing data as well as data‐driven models to enable AI innovation. A broad range of domain areas is covered by the funded efforts, spanning across healthcare and medicine, to climate change and disaster, and civil/built infrastructure. The projects are addressing sharing of open as well as sensitive/private data. In September 2021, six of the eighteen projects described here were selected for phase II of the program, as noted in this article.
An incremental version of the ELMVIS+ method is proposed in this paper. It iteratively selects a few best fitting data samples from a large pool, and adds them to the model. The method keeps high speed of ELMVIS+ while allowing for much larger possible sample pools due to lower memory requirements. The extension is useful for reaching a better local optimum with greedy optimization of ELMVIS, and the data structure can be specified in semi-supervised optimization. The major new application of incremental ELMVIS is not to visualization, but to a general dataset processing. The method is capable of learning dependencies from non-organized unsupervised data -- either reconstructing a shuffled dataset, or learning dependencies in complex high-dimensional space. The results are interesting and promising, although there is space for improvements.
'%3e%3cg id='b'%3e%3cpath id='a' stroke='%23000' stroke-width='2' d='M-6-25H6m-12 3H6m-12 3H6'/%3e%3cuse xlink:href='%23a' y='44'/%3e%3c/g%3e%3cpath stroke='%23fff' d='M0 17v10'/%3e%3ccircle r='12' fill='%23cd2e3a'/%3e%3cpath fill='%230047a0' d='M0-12A6 6 0 0 0 0 0a6 6 0 0 1 0 12 12 12 0 0 1 0-24z'/%3e%3c/g%3e%3cg transform='rotate(-123.69)'%3e%3cuse xlink:href='%23b'/%3e%3cpath stroke='%23fff' d='M0-23.5v3M0 17v3.5m0 3v3'/%3e%3c/g%3e%3c/svg%3e)