While neural network approaches have made significant strides in resolving classical signal processing problems, it is often the case that hybrid approaches that draw insight from both signal processing and neural networks produce more complete solutions. In this paper, we present a hybrid classical digital signal processing/deep neural network (DSP/DNN) approach to source separation (SS) highlighting the theoretical link between variational autoencoder and classical approaches to SS. We propose a system that transforms the single channel under-determined SS task to an equivalent multichannel over-determined SS problem in a properly designed latent space. The separation task in the latent space is treated as finding a variational block-wise disentangled representation of the mixture. We show empirically, that the design choices and the variational formulation of the task at hand motivated by the classical signal processing theoretical results lead to robustness to unseen out-of-distribution data and reduction of the overfitting risk. To address the resulting permutation issue we explicitly incorporate a novel differentiable permutation loss function and augment the model with a memory mechanism to keep track of the statistics of the individual sources.
相關內容
Hybrid non-orthogonal multiple access (H-NOMA) has recently received significant attention as a general framework of multiple access, where both conventional orthogonal multiple access (OMA) and pure NOMA are its special cases. This paper focuses on the application of H-NOMA to ambient Internet of Things (IoT) with energy-constrained devices, where a new backscatter communication (BackCom) assisted H-NOMA uplink scheme is developed. Resource allocation for H-NOMA uplink transmission is also considered, where an overall power minimization problem is formulated. Insightful understandings for the key features of BackCom assisted H-NOMA and its difference from conventional H-NOMA are illustrated by developing analytical results for the two-user special case. For the general multi-user scenario, two algorithms, one based on the branch-bound (BB) principle and the other based on successive convex approximation (SCA), are developed to realize different tradeoffs between the system performance and complexity. The numerical results are also provided to verify the accuracy of the developed analytical results and demonstrate the performance gain of H-NOMA over OMA.
Movable antenna (MA) is a promising technology to improve wireless communication performance by varying the antenna position in a given finite area at the transceivers to create more favorable channel conditions. In this paper, we investigate the MA-enhanced multiple-access channel (MAC) for the uplink transmission from multiple users each equipped with a single MA to a base station (BS) with a fixed-position antenna (FPA) array. A field-response based channel model is used to characterize the multi-path channel between the antenna array of the BS and each user's MA with a flexible position. To evaluate the MAC performance gain provided by MAs, we formulate an optimization problem for minimizing the total transmit power of users, subject to a minimum-achievable-rate requirement for each user, where the positions of MAs and the transmit powers of users, as well as the receive combining matrix of the BS are jointly optimized. To solve this non-convex optimization problem involving intricately coupled variables, we develop two algorithms based on zero-forcing (ZF) and minimum mean square error (MMSE) combining methods, respectively. Specifically, for each algorithm, the combining matrix of the BS and the total transmit power of users are expressed as a function of the MAs' position vectors, which are then optimized by using the proposed multi-directional descent (MDD) framework. It is shown that the proposed ZF-based and MMSE-based MDD algorithms can converge to high-quality suboptimal solutions with low computational complexities. Simulation results demonstrate that the proposed solutions for MA-enhanced multiple access systems can significantly decrease the total transmit power of users as compared to conventional FPA systems employing antenna selection under both perfect and imperfect field-response information.
In the rapidly evolving landscape of artificial intelligence, multimodal learning systems (MMLS) have gained traction for their ability to process and integrate information from diverse modality inputs. Their expanding use in vital sectors such as healthcare has made safety assurance a critical concern. However, the absence of systematic research into their safety is a significant barrier to progress in this field. To bridge the gap, we present the first taxonomy that systematically categorizes and assesses MMLS safety. This taxonomy is structured around four fundamental pillars that are critical to ensuring the safety of MMLS: robustness, alignment, monitoring, and controllability. Leveraging this taxonomy, we review existing methodologies, benchmarks, and the current state of research, while also pinpointing the principal limitations and gaps in knowledge. Finally, we discuss unique challenges in MMLS safety. In illuminating these challenges, we aim to pave the way for future research, proposing potential directions that could lead to significant advancements in the safety protocols of MMLS.
Data-driven optimization models have the potential to significantly improve hospital capacity management, particularly during demand surges, when effective allocation of capacity is most critical and challenging. However, integrating models into existing processes in a way that provides value requires recognizing that hospital administrators are ultimately responsible for making capacity management decisions, and carefully building trustworthy and accessible tools for them. In this study, we develop an interactive, user-friendly, electronic dashboard for informing hospital capacity management decisions during surge periods. The dashboard integrates real-time hospital data, predictive analytics, and optimization models. It allows hospital administrators to interactively customize parameters, enabling them to explore a range of scenarios, and provides real-time updates on recommended optimal decisions. The dashboard was created through a participatory design process, involving hospital administrators in the development team to ensure practical utility, trustworthiness, transparency, explainability, and usability. We successfully deployed our dashboard within the Johns Hopkins Health System during the height of the COVID-19 pandemic, addressing the increased need for tools to inform hospital capacity management. It was used on a daily basis, with results regularly communicated to hospital leadership. This study demonstrates the practical application of a prospective, data-driven, interactive decision-support tool for hospital system capacity management.
Experimentation in practical, end-to-end (E2E) next-generation networks deployments is becoming increasingly prevalent and significant in the realm of modern networking and wireless communications research. The prevalence of fifth-generation technology (5G) testbeds and the emergence of developing networks systems, for the purposes of research and testing, focus on the capabilities and features of analytics, intelligence, and automated management using novel testbed designs and architectures, ranging from simple simulations and setups to complex networking systems; however, with the ever-demanding application requirements for modern and future networks, 5G-and-beyond (denoted as 5G+) testbed experimentation can be useful in assessing the creation of large-scale network infrastructures that are capable of supporting E2E virtualized mobile network services. To this end, this paper presents a functional, modular E2E 5G+ system, complete with the integration of a Radio Access Network (RAN) and handling the connection of User Equipment (UE) in real-world scenarios. As well, this paper assesses and evaluates the effectiveness of emulating full network functionalities and capabilities, including a complete description of user-plane data, from UE registrations to communications sequences, and leads to the presentation of a future outlook in powering new experimentation for 6G and next-generation networks.
This thesis investigates how natural language understanding and generation with transformer models can benefit from grounding the models with knowledge representations and addresses the following key research questions: (i) Can knowledge of entities extend its benefits beyond entity-centric tasks, such as entity linking? (ii) How can we faithfully and effectively extract such structured knowledge from raw text, especially noisy web text? (iii) How do other types of knowledge, beyond structured knowledge, contribute to improving NLP tasks? Studies in this thesis find that incorporating relevant and up-to-date knowledge of entities benefits fake news detection, and entity-focused code-switching significantly enhances zero-shot cross-lingual transfer on entity-centric tasks. In terms of effective and faithful approaches to extracting structured knowledge, it is observed that integrating negative examples and training with entity planning significantly improves performance. Additionally, it is established that other general forms of knowledge, such as parametric and distilled knowledge, enhance multimodal and multilingual knowledge-intensive tasks. This research shows the tangible benefits of diverse knowledge integration and motivates further exploration in this direction.
A mainstream type of current self-supervised learning methods pursues a general-purpose representation that can be well transferred to downstream tasks, typically by optimizing on a given pretext task such as instance discrimination. In this work, we argue that existing pretext tasks inevitably introduce biases into the learned representation, which in turn leads to biased transfer performance on various downstream tasks. To cope with this issue, we propose Maximum Entropy Coding (MEC), a more principled objective that explicitly optimizes on the structure of the representation, so that the learned representation is less biased and thus generalizes better to unseen downstream tasks. Inspired by the principle of maximum entropy in information theory, we hypothesize that a generalizable representation should be the one that admits the maximum entropy among all plausible representations. To make the objective end-to-end trainable, we propose to leverage the minimal coding length in lossy data coding as a computationally tractable surrogate for the entropy, and further derive a scalable reformulation of the objective that allows fast computation. Extensive experiments demonstrate that MEC learns a more generalizable representation than previous methods based on specific pretext tasks. It achieves state-of-the-art performance consistently on various downstream tasks, including not only ImageNet linear probe, but also semi-supervised classification, object detection, instance segmentation, and object tracking. Interestingly, we show that existing batch-wise and feature-wise self-supervised objectives could be seen equivalent to low-order approximations of MEC. Code and pre-trained models are available at //github.com/xinliu20/MEC.
The key challenge of image manipulation detection is how to learn generalizable features that are sensitive to manipulations in novel data, whilst specific to prevent false alarms on authentic images. Current research emphasizes the sensitivity, with the specificity overlooked. In this paper we address both aspects by multi-view feature learning and multi-scale supervision. By exploiting noise distribution and boundary artifact surrounding tampered regions, the former aims to learn semantic-agnostic and thus more generalizable features. The latter allows us to learn from authentic images which are nontrivial to be taken into account by current semantic segmentation network based methods. Our thoughts are realized by a new network which we term MVSS-Net. Extensive experiments on five benchmark sets justify the viability of MVSS-Net for both pixel-level and image-level manipulation detection.
Approaches based on deep neural networks have achieved striking performance when testing data and training data share similar distribution, but can significantly fail otherwise. Therefore, eliminating the impact of distribution shifts between training and testing data is crucial for building performance-promising deep models. Conventional methods assume either the known heterogeneity of training data (e.g. domain labels) or the approximately equal capacities of different domains. In this paper, we consider a more challenging case where neither of the above assumptions holds. We propose to address this problem by removing the dependencies between features via learning weights for training samples, which helps deep models get rid of spurious correlations and, in turn, concentrate more on the true connection between discriminative features and labels. Extensive experiments clearly demonstrate the effectiveness of our method on multiple distribution generalization benchmarks compared with state-of-the-art counterparts. Through extensive experiments on distribution generalization benchmarks including PACS, VLCS, MNIST-M, and NICO, we show the effectiveness of our method compared with state-of-the-art counterparts.
Learning latent representations of nodes in graphs is an important and ubiquitous task with widespread applications such as link prediction, node classification, and graph visualization. Previous methods on graph representation learning mainly focus on static graphs, however, many real-world graphs are dynamic and evolve over time. In this paper, we present Dynamic Self-Attention Network (DySAT), a novel neural architecture that operates on dynamic graphs and learns node representations that capture both structural properties and temporal evolutionary patterns. Specifically, DySAT computes node representations by jointly employing self-attention layers along two dimensions: structural neighborhood and temporal dynamics. We conduct link prediction experiments on two classes of graphs: communication networks and bipartite rating networks. Our experimental results show that DySAT has a significant performance gain over several different state-of-the-art graph embedding baselines.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191