Search

Additional file 1.

Supplemental material, sj-docx-1-tag-10.1177_17562848231155023 for An explainable artificial intelligence system for diagnosing Helicobacter Pylori infection under endoscopy: a case–control study by Mengjiao Zhang, Jie Pan, Jiejun Lin, Ming Xu, Lihui Zhang, Renduo Shang, Liwen Yao, Yanxia Li, Wei Zhou, Yunchao Deng, Zehua Dong, Yijie Zhu, Xiao Tao, Lianlian Wu and Honggang Yu in Therapeutic Advances in Gastroenterology

Additional file 1.

Building a scalable and real-time recommendation system is vital for many businesses driven by time-sensitive customer feedback, such as short-videos ranking or online ads. Despite the ubiquitous adoption of production-scale deep learning frameworks like TensorFlow or PyTorch, these general-purpose frameworks fall short of business demands in recommendation scenarios for various reasons: on one hand, tweaking systems based on static parameters and dense computations for recommendation with dynamic and sparse features is detrimental to model quality; on the other hand, such frameworks are designed with batch-training stage and serving stage completely separated, preventing the model from interacting with customer feedback in real-time. These issues led us to reexamine traditional approaches and explore radically different design choices. In this paper, we present Monolith, a system tailored for online training. Our design has been driven by observations of our application workloads and production environment that reflects a marked departure from other recommendations systems. Our contributions are manifold: first, we crafted a collisionless embedding table with optimizations such as expirable embeddings and frequency filtering to reduce its memory footprint; second, we provide an production-ready online training architecture with high fault-tolerance; finally, we proved that system reliability could be traded-off for real-time learning. Monolith has successfully landed in the BytePlus Recommend product., Comment: ORSUM@ACM RecSys 2022

The click-through rate (CTR) prediction task is to predict whether a user will click on the recommended item. As mind-boggling amounts of data are produced online daily, accelerating CTR prediction model training is critical to ensuring an up-to-date model and reducing the training cost. One approach to increase the training speed is to apply large batch training. However, as shown in computer vision and natural language processing tasks, training with a large batch easily suffers from the loss of accuracy. Our experiments show that previous scaling rules fail in the training of CTR prediction neural networks. To tackle this problem, we first theoretically show that different frequencies of ids make it challenging to scale hyperparameters when scaling the batch size. To stabilize the training process in a large batch size setting, we develop the adaptive Column-wise Clipping (CowClip). It enables an easy and effective scaling rule for the embeddings, which keeps the learning rate unchanged and scales the L2 loss. We conduct extensive experiments with four CTR prediction networks on two real-world datasets and successfully scaled 128 times the original batch size without accuracy loss. In particular, for CTR prediction model DeepFM training on the Criteo dataset, our optimization framework enlarges the batch size from 1K to 128K with over 0.1% AUC improvement and reduces training time from 12 hours to 10 minutes on a single V100 GPU. Our code locates at https://github.com/bytedance/LargeBatchCTR., Comment: AAAI 2023

Rayleigh scattering of the cosmic microwave background (CMB) by neutral hydrogen shortly after recombination leaves frequency-dependent imprints on intensity and polarization fluctuations. High signal-to-noise observations of CMB Rayleigh scattering would provide additional insight into the physics of recombination, including greater constraining power for parameters like the primordial helium fraction, the light relic density, and the sum of neutrino masses. However, such a measurement of CMB Rayleigh scattering is challenging due to the presence of astrophysical foregrounds, which are more intense at the high frequencies, where the effects of Rayleigh scattering are most prominent. Here we forecast the detectability of CMB Rayleigh scattering including foreground removal using blind internal linear combination methods for a set of near-future surveys. We show that atmospheric effects for ground-based observatories and astrophysical foregrounds pose a significant hindrance to detecting CMB Rayleigh scattering with experiments planned for this decade, though a high-significance measurement should be possible with a future CMB satellite., Comment: 14 pages, 12 figures

With the rapid development of the location service industry, various positioning technologies have emerged. Recently, the mainstream indoor positioning technologies include WiFi positioning and Bluetooth positioning. Various positioning methods have their own advantages and disadvantages due to their different positioning technologies. This paper proposes an indoor positioning method based on WiFi, Bluetooth and PDR fusion positioning. Firstly, WiFi positioning and Bluetooth positioning are achieved by improving the weighted centroid method. The WiFi and Bluetooth positioning are integrated, and the positioning result is integrated by weight adaptive constraint, which solves the problem of WiFi signal instability. The fusion positioning result and PDR positioning fusion are used to achieve fusion positioning through UKF, which solves the problem of large cumulative error in PDR positioning. The experiment proves that the WiFi, Bluetooth and PDR fusion positioning results are higher than the positioning accuracy of the individual positioning, which solves the problem that the WiFi positioning signal is unstable and the PDR cumulative error is large.

The guide rod tandem suspension was divided into rigid parts and flexible parts by the analysis and simplification of its structure. In ADAMS, the model of rigid parts was built, and the model of flexible parts was built by the discrete beam element method. A integrated dynamic model of the guide rod tandem suspension was assembled by reasonable constraints between rigid parts and flexible parts. In order to validate the model, road vertical displacement motion actuators of four wheels were applied to simulate parallel wheel travel, pitch, roll and pitch-roll cases of the guide rod tandem suspension, the motion tracks of the suspension for four cases validated the reasonability of model.