Federated learning (FL) has been recognized as a privacy-preserving distributed machine learning paradigm that enables knowledge sharing among various heterogeneous artificial intelligence (AIoT) devices through centralized global model aggregation. FL suffers from model inaccuracy and slow convergence due to the model heterogeneity of the AIoT devices involved. Although various existing methods try to solve the bottleneck of the model heterogeneity problem, most of them improve the accuracy of heterogeneous models in a coarse-grained manner, which makes it still a great challenge to deploy large-scale AIoT devices. To alleviate the negative impact of this problem and take full advantage of the diversity of each heterogeneous model, we propose an efficient framework named HierarchyFL, which uses a small amount of public data for efficient and scalable knowledge across a variety of differently structured models. By using self-distillation and our proposed ensemble library, each hierarchical model can intelligently learn from each other on cloud servers. Experimental results on various well-known datasets show that HierarchyFL can not only maximize the knowledge sharing among various heterogeneous models in large-scale AIoT systems, but also greatly improve the model performance of each involved heterogeneous AIoT device.

The role of mobile cameras increased dramatically over the past few years, leading to more and more research in automatic image quality enhancement and RAW photo processing. In this Mobile AI challenge, the target was to develop an efficient end-to-end AI-based image signal processing (ISP) pipeline replacing the standard mobile ISPs that can run on modern smartphone GPUs using TensorFlow Lite. The participants were provided with a large-scale Fujifilm UltraISP dataset consisting of thousands of paired photos captured with a normal mobile camera sensor and a professional 102MP medium-format FujiFilm GFX100 camera. The runtime of the resulting models was evaluated on the Snapdragon's 8 Gen 1 GPU that provides excellent acceleration results for the majority of common deep learning ops. The proposed solutions are compatible with all recent mobile GPUs, being able to process Full HD photos in less than 20-50 milliseconds while achieving high fidelity results. A detailed description of all models developed in this challenge is provided in this paper.

在联合学习等协作学习环境中，好奇的疗程可能是诚实的，但正在通过推理攻击试图通过推断攻击推断其他方的私人数据，而恶意缔约方可能会通过后门攻击操纵学习过程。但是，大多数现有的作品只考虑通过样本（HFL）划分数据的联合学习场景。特征分区联合学习（VFL）可以是许多真实应用程序中的另一个重要方案。当攻击者和防守者无法访问其他参与者的功能或模型参数时，这种情况下的攻击和防御尤其挑战。以前的作品仅显示了可以从每个样本渐变重建私有标签。在本文中，我们首先表明，只有批量平均梯度被揭示时，可以重建私人标签，这是针对常见的推定。此外，我们表明VFL中的被动派对甚至可以通过梯度替换攻击将其相应的标签用目标标签替换为目标标签。为了防御第一次攻击，我们介绍了一种基于AutoEncoder和熵正则化的混乱自动化器（CoAE）的新技术。我们证明，与现有方法相比，这种技术可以成功阻止标签推理攻击，同时损害较少的主要任务准确性。我们的COAE技术在捍卫梯度替代后门攻击方面也有效，使其成为一个普遍和实用的防御策略，没有改变原来的VFL协议。我们展示了我们双方和多方VFL设置下的方法的有效性。据我们所知，这是第一次处理特征分区联合学习框架中的标签推理和后门攻击的第一个系统研究。

Label noise is ubiquitous in various machine learning scenarios such as self-labeling with model predictions and erroneous data annotation. Many existing approaches are based on heuristics such as sample losses, which might not be flexible enough to achieve optimal solutions. Meta learning based methods address this issue by learning a data selection function, but can be hard to optimize. In light of these pros and cons, we propose Selection-Enhanced Noisy label Training (SENT) that does not rely on meta learning while having the flexibility of being data-driven. SENT transfers the noise distribution to a clean set and trains a model to distinguish noisy labels from clean ones using model-based features. Empirically, on a wide range of tasks including text classification and speech recognition, SENT improves performance over strong baselines under the settings of self-training and label corruption.

In this paper, we propose a large-scale language pre-training for text GENeration using dIffusion modEl, which is named GENIE. GENIE is a pre-training sequence-to-sequence text generation model which combines Transformer and diffusion. The diffusion model accepts the latent information from the encoder, which is used to guide the denoising of the current time step. After multiple such denoise iterations, the diffusion model can restore the Gaussian noise to the diverse output text which is controlled by the input text. Moreover, such architecture design also allows us to adopt large scale pre-training on the GENIE. We propose a novel pre-training method named continuous paragraph denoise based on the characteristics of the diffusion model. Extensive experiments on the XSum, CNN/DailyMail, and Gigaword benchmarks shows that GENIE can achieves comparable performance with various strong baselines, especially after pre-training, the generation quality of GENIE is greatly improved. We have also conduct a lot of experiments on the generation diversity and parameter impact of GENIE. The code for GENIE will be made publicly available.

In recent years, neural image compression (NIC) algorithms have shown powerful coding performance. However, most of them are not adaptive to the image content. Although several content adaptive methods have been proposed by updating the encoder-side components, the adaptability of both latents and the decoder is not well exploited. In this work, we propose a new NIC framework that improves the content adaptability on both latents and the decoder. Specifically, to remove redundancy in the latents, our content adaptive channel dropping (CACD) method automatically selects the optimal quality levels for the latents spatially and drops the redundant channels. Additionally, we propose the content adaptive feature transformation (CAFT) method to improve decoder-side content adaptability by extracting the characteristic information of the image content, which is then used to transform the features in the decoder side. Experimental results demonstrate that our proposed methods with the encoder-side updating algorithm achieve the state-of-the-art performance.

Conventional cameras capture image irradiance on a sensor and convert it to RGB images using an image signal processor (ISP). The images can then be used for photography or visual computing tasks in a variety of applications, such as public safety surveillance and autonomous driving. One can argue that since RAW images contain all the captured information, the conversion of RAW to RGB using an ISP is not necessary for visual computing. In this paper, we propose a novel $\rho$-Vision framework to perform high-level semantic understanding and low-level compression using RAW images without the ISP subsystem used for decades. Considering the scarcity of available RAW image datasets, we first develop an unpaired CycleR2R network based on unsupervised CycleGAN to train modular unrolled ISP and inverse ISP (invISP) models using unpaired RAW and RGB images. We can then flexibly generate simulated RAW images (simRAW) using any existing RGB image dataset and finetune different models originally trained for the RGB domain to process real-world camera RAW images. We demonstrate object detection and image compression capabilities in RAW-domain using RAW-domain YOLOv3 and RAW image compressor (RIC) on snapshots from various cameras. Quantitative results reveal that RAW-domain task inference provides better detection accuracy and compression compared to RGB-domain processing. Furthermore, the proposed \r{ho}-Vision generalizes across various camera sensors and different task-specific models. Additional advantages of the proposed $\rho$-Vision that eliminates the ISP are the potential reductions in computations and processing times.

Homography estimation is a basic computer vision task, which aims to obtain the transformation from multi-view images for image alignment. Unsupervised learning homography estimation trains a convolution neural network for feature extraction and transformation matrix regression. While the state-of-theart homography method is based on convolution neural networks, few work focuses on transformer which shows superiority in highlevel vision tasks. In this paper, we propose a strong-baseline model based on the Swin Transformer, which combines convolution neural network for local features and transformer module for global features. Moreover, a cross non-local layer is introduced to search the matched features within the feature maps coarsely. In the homography regression stage, we adopt an attention layer for the channels of correlation volume, which can drop out some weak correlation feature points. The experiment shows that in 8 Degree-of-Freedoms(DOFs) homography estimation our method overperforms the state-of-the-art method.

We propose a PiggyBack, a Visual Question Answering platform that allows users to apply the state-of-the-art visual-language pretrained models easily. The PiggyBack supports the full stack of visual question answering tasks, specifically data processing, model fine-tuning, and result visualisation. We integrate visual-language models, pretrained by HuggingFace, an open-source API platform of deep learning technologies; however, it cannot be runnable without programming skills or deep learning understanding. Hence, our PiggyBack supports an easy-to-use browser-based user interface with several deep learning visual language pretrained models for general users and domain experts. The PiggyBack includes the following benefits: Free availability under the MIT License, Portability due to web-based and thus runs on almost any platform, A comprehensive data creation and processing technique, and ease of use on deep learning-based visual language pretrained models. The demo video is available on YouTube and can be found at https://youtu.be/iz44RZ1lF4s.

With the continued integration of autonomous vehicles (AVs) into public roads, a mixed traffic environment with large-scale human-driven vehicles (HVs) and AVs interactions is imminent. In challenging traffic scenarios, such as emergency braking, it is crucial to account for the reactive and uncertain behavior of HVs when developing control strategies for AVs. This paper studies the safe control of a platoon of AVs interacting with a human-driven vehicle in longitudinal car-following scenarios. We first propose the use of a model that combines a first-principles model (nominal model) with a Gaussian process (GP) learning-based component for predicting behaviors of the human-driven vehicle when it interacts with AVs. The modeling accuracy of the proposed method shows a $9\%$ reduction in root mean square error (RMSE) in predicting a HV's velocity compared to the nominal model. Exploiting the properties of this model, we design a model predictive control (MPC) strategy for a platoon of AVs to ensure a safe distance between each vehicle, as well as a (probabilistic) safety of the human-driven car following the platoon. Compared to a baseline MPC that uses only a nominal model for HVs, our method achieves better velocity-tracking performance for the autonomous vehicle platoon and more robust constraint satisfaction control for a platoon of mixed vehicles system. Simulation studies demonstrate a $4.2\%$ decrease in the control cost and an approximate $1m$ increase in the minimum distance between autonomous and human-driven vehicles to better guarantee safety in challenging traffic scenarios.