Solving the camera-to-robot pose is a fundamental requirement for vision-based robot control, and is a process that takes considerable effort and cares to make accurate. Traditional approaches require modification of the robot via markers, and subsequent deep learning approaches enabled markerless feature extraction. Mainstream deep learning methods only use synthetic data and rely on Domain Randomization to fill the sim-to-real gap, because acquiring the 3D annotation is labor-intensive. In this work, we go beyond the limitation of 3D annotations for real-world data. We propose an end-to-end pose estimation framework that is capable of online camera-to-robot calibration and a self-supervised training method to scale the training to unlabeled real-world data. Our framework combines deep learning and geometric vision for solving the robot pose, and the pipeline is fully differentiable. To train the Camera-to-Robot Pose Estimation Network (CtRNet), we leverage foreground segmentation and differentiable rendering for image-level self-supervision. The pose prediction is visualized through a renderer and the image loss with the input image is back-propagated to train the neural network. Our experimental results on two public real datasets confirm the effectiveness of our approach over existing works. We also integrate our framework into a visual servoing system to demonstrate the promise of real-time precise robot pose estimation for automation tasks.
Federated learning, which allows multiple client devices in a network to jointly train a machine learning model without direct exposure of clients' data, is an emerging distributed learning technique due to its nature of privacy preservation. However, it has been found that models trained with federated learning usually have worse performance than their counterparts trained in the standard centralized learning mode, especially when the training data is imbalanced. In the context of federated learning, data imbalance may occur either locally one one client device, or globally across many devices. The complexity of different types of data imbalance has posed challenges to the development of federated learning technique, especially considering the need of relieving data imbalance issue and preserving data privacy at the same time. Therefore, in the literature, many attempts have been made to handle class imbalance in federated learning. In this paper, we present a detailed review of recent advancements along this line. We first introduce various types of class imbalance in federated learning, after which we review existing methods for estimating the extent of class imbalance without the need of knowing the actual data to preserve data privacy. After that, we discuss existing methods for handling class imbalance in FL, where the advantages and disadvantages of the these approaches are discussed. We also summarize common evaluation metrics for class imbalanced tasks, and point out potential future directions.
The prompt estimation of traffic incident impacts can guide commuters in their trip planning and improve the resilience of transportation agencies' decision-making on resilience. However, it is more challenging than node-level and graph-level forecasting tasks, as it requires extracting the anomaly subgraph or sub-time-series from dynamic graphs. In this paper, we propose DG-Trans, a novel traffic incident impact prediction framework, to foresee the impact of traffic incidents through dynamic graph learning. The proposed framework contains a dual-level spatial transformer and an importance-score-based temporal transformer, and the performance of this framework is justified by two newly constructed benchmark datasets. The dual-level spatial transformer removes unnecessary edges between nodes to isolate the affected subgraph from the other nodes. Meanwhile, the importance-score-based temporal transformer identifies abnormal changes in node features, causing the predictions to rely more on measurement changes after the incident occurs. Therefore, DG-Trans is equipped with dual abilities that extract spatiotemporal dependency and identify anomaly nodes affected by incidents while removing noise introduced by benign nodes. Extensive experiments on real-world datasets verify that DG-Trans outperforms the existing state-of-the-art methods, especially in extracting spatiotemporal dependency patterns and predicting traffic accident impacts. It offers promising potential for traffic incident management systems.
We study speech enhancement using deep learning (DL) for virtual meetings on cellular devices, where transmitted speech has background noise and transmission loss that affects speech quality. Since the Deep Noise Suppression (DNS) Challenge dataset does not contain practical disturbance, we collect a transmitted DNS (t-DNS) dataset using Zoom Meetings over T-Mobile network. We select two baseline models: Demucs and FullSubNet. The Demucs is an end-to-end model that takes time-domain inputs and outputs time-domain denoised speech, and the FullSubNet takes time-frequency-domain inputs and outputs the energy ratio of the target speech in the inputs. The goal of this project is to enhance the speech transmitted over the cellular networks using deep learning models.
For tabular data sets, we explore data and model distillation, as well as data denoising. These techniques improve both gradient-boosting models and a specialized DNN architecture. While gradient boosting is known to outperform DNNs on tabular data, we close the gap for datasets with 100K+ rows and give DNNs an advantage on small data sets. We extend these results with input-data distillation and optimized ensembling to help DNN performance match or exceed that of gradient boosting. As a theoretical justification of our practical method, we prove its equivalence to classical cross-entropy knowledge distillation. We also qualitatively explain the superiority of DNN ensembles over XGBoost on small data sets. For an industry end-to-end real-time ML platform with 4M production inferences per second, we develop a model-training workflow based on data sampling that distills ensembles of models into a single gradient-boosting model favored for high-performance real-time inference, without performance loss. Empirical evaluation shows that the proposed combination of methods consistently improves model accuracy over prior best models across several production applications deployed worldwide.
Machine Learning (ML)-powered apps are used in pervasive devices such as phones, tablets, smartwatches and IoT devices. Recent advances in collaborative, distributed ML such as Federated Learning (FL) attempt to solve privacy concerns of users and data owners, and thus used by tech industry leaders such as Google, Facebook and Apple. However, FL systems and models are still vulnerable to adversarial membership and attribute inferences and model poisoning attacks, especially in FL-as-a-Service ecosystems recently proposed, which can enable attackers to access multiple ML-powered apps. In this work, we focus on the recently proposed Sponge attack: It is designed to soak up energy consumed while executing inference (not training) of ML model, without hampering the classifier's performance. Recent work has shown sponge attacks on ASCI-enabled GPUs can potentially escalate the power consumption and inference time. For the first time, in this work, we investigate this attack in the mobile setting and measure the effect it can have on ML models running inside apps on mobile devices.
Deep artificial neural networks (ANNs) play a major role in modeling the visual pathways of primate and rodent. However, they highly simplify the computational properties of neurons compared to their biological counterparts. Instead, Spiking Neural Networks (SNNs) are more biologically plausible models since spiking neurons encode information with time sequences of spikes, just like biological neurons do. However, there is a lack of studies on visual pathways with deep SNNs models. In this study, we model the visual cortex with deep SNNs for the first time, and also with a wide range of state-of-the-art deep CNNs and ViTs for comparison. Using three similarity metrics, we conduct neural representation similarity experiments on three neural datasets collected from two species under three types of stimuli. Based on extensive similarity analyses, we further investigate the functional hierarchy and mechanisms across species. Almost all similarity scores of SNNs are higher than their counterparts of CNNs with an average of 6.6%. Depths of the layers with the highest similarity scores exhibit little differences across mouse cortical regions, but vary significantly across macaque regions, suggesting that the visual processing structure of mice is more regionally homogeneous than that of macaques. Besides, the multi-branch structures observed in some top mouse brain-like neural networks provide computational evidence of parallel processing streams in mice, and the different performance in fitting macaque neural representations under different stimuli exhibits the functional specialization of information processing in macaques. Taken together, our study demonstrates that SNNs could serve as promising candidates to better model and explain the functional hierarchy and mechanisms of the visual system.
Adversarial machine learning (AML) studies the adversarial phenomenon of machine learning, which may make inconsistent or unexpected predictions with humans. Some paradigms have been recently developed to explore this adversarial phenomenon occurring at different stages of a machine learning system, such as training-time adversarial attack (i.e., backdoor attack), deployment-time adversarial attack (i.e., weight attack), and inference-time adversarial attack (i.e., adversarial example). However, although these paradigms share a common goal, their developments are almost independent, and there is still no big picture of AML. In this work, we aim to provide a unified perspective to the AML community to systematically review the overall progress of this field. We firstly provide a general definition about AML, and then propose a unified mathematical framework to covering existing attack paradigms. According to the proposed unified framework, we can not only clearly figure out the connections and differences among these paradigms, but also systematically categorize and review existing works in each paradigm.
Automatic underwater vehicle hull Design optimization is a complex engineering process for generating a UUV hull with optimized properties on a given requirement. First, it involves the integration of involved computationally complex engineering simulation tools. Second, it needs integration of a sample efficient optimization framework with the integrated toolchain. To this end, we integrated the CAD tool called FreeCAD with CFD tool openFoam for automatic design evaluation. For optimization, we chose Bayesian optimization (BO), which is a well-known technique developed for optimizing time-consuming expensive engineering simulations and has proven to be very sample efficient in a variety of problems, including hyper-parameter tuning and experimental design. During the optimization process, we can handle infeasible design as constraints integrated into the optimization process. By integrating domain-specific toolchain with AI-based optimization, we executed the automatic design optimization of underwater vehicle hull design. For empirical evaluation, we took two different use cases of real-world underwater vehicle design to validate the execution of our tool.
With the increased interest in immersive experiences, point cloud came to birth and was widely adopted as the first choice to represent 3D media. Besides several distortions that could affect the 3D content spanning from acquisition to rendering, efficient transmission of such volumetric content over traditional communication systems stands at the expense of the delivered perceptual quality. To estimate the magnitude of such degradation, employing quality metrics became an inevitable solution. In this work, we propose a novel deep-based no-reference quality metric that operates directly on the whole point cloud without requiring extensive pre-processing, enabling real-time evaluation over both transmission and rendering levels. To do so, we use a novel model design consisting primarily of cross and self-attention layers, in order to learn the best set of local semantic affinities while keeping the best combination of geometry and color information in multiple levels from basic features extraction to deep representation modeling.