A large and growing amount of speech content in real-life scenarios is being recorded on common consumer devices in uncontrolled environments, resulting in degraded speech quality. Transforming such low-quality device-degraded speech into high-qualit
y speech is a goal of speech enhancement (SE). This paper introduces a new speech dataset, DDS, to facilitate the research on SE. DDS provides aligned parallel recordings of high-quality speech (recorded in professional studios) and a number
Neural network applications have become popular in both enterprise and personal settings. Network solutions are tuned meticulously for each task, and designs that can robustly resolve queries end up in high demand. As the commercial value of accurate
and performant machine learning models increases, so too does the demand to protect neural architectures as confidential investments. We explore the vulnerability of neural networks deployed as black boxes across accelerated hardware through electromagnetic side channels. We examine the magnetic flux emanating from a graphics processing units power cable, as acquired by a cheap $3 induction sensor, and find that this signal betrays the detailed topology and hyperparameters of a black-box neural network model. The attack acquires the magnetic signal for one query with unknown input values, but known input dimensions. The network reconstruction is possible due to the modular layer sequence in which deep neural networks are evaluated. We find that each layer components evaluation produces an identifiable magnetic signal signature, from which layer topology, width, function type, and sequence order can be inferred using a suitably trained classifier and a joint consistency optimization based on integer programming. We study the extent to which network specifications can be recovered, and consider metrics for comparing network similarity. We demonstrate the potential accuracy of this side channel attack in recovering the details for a broad range of network architectures, including random designs. We consider applications that may exploit this novel side channel exposure, such as adversarial transfer attacks. In response, we discuss countermeasures to protect against our method and other similar snooping techniques.
Recently, the Vision Transformer (ViT) has shown impressive performance on high-level and low-level vision tasks. In this paper, we propose a new ViT architecture, named Hybrid Local-Global Vision Transformer (HyLoG-ViT), for single image dehazing. T
he HyLoG-ViT block consists of two paths, the local ViT path and the global ViT path, which are used to capture local and global dependencies. The hybrid features are fused via convolution layers. As a result, the HyLoG-ViT reduces the computational complexity and introduces locality in the networks. Then, the HyLoG-ViT blocks are incorporated within our dehazing networks, which jointly learn the intrinsic image decomposition and image dehazing. Specifically, the network consists of one shared encoder and three decoders for reflectance prediction, shading prediction, and haze-free image generation. The tasks of reflectance and shading prediction can produce meaningful intermediate features that can serve as complementary features for haze-free image generation. To effectively aggregate the complementary features, we propose a complementary features selection module (CFSM) to select the useful ones for image dehazing. Extensive experiments on homogeneous, non-homogeneous, and nighttime dehazing tasks reveal that our proposed Transformer-based dehazing network can achieve comparable or even better performance than CNNs-based dehazing models.
Mobile Edge Caching is a promising technique to enhance the content delivery quality and reduce the backhaul link congestion, by storing popular content at the network edge or mobile devices (e.g. base stations and smartphones) that are proximate to
content requesters. In this work, we study a novel mobile edge caching framework, which enables mobile devices to cache and share popular contents with each other via device-to-device (D2D) links. We are interested in the following incentive problem of mobile device users: whether and which users are willing to cache and share what contents, taking the user mobility and cost/reward into consideration. The problem is challenging in a large-scale network with a large number of users. We introduce the evolutionary game theory, an effective tool for analyzing large-scale dynamic systems, to analyze the mobile users content caching and sharing strategies. Specifically, we first derive the users best caching and sharing strategies, and then analyze how these best strategies change dynamically over time, based on which we further characterize the system equilibrium systematically. Simulation results show that the proposed caching scheme outperforms the existing schemes in terms of the total transmission cost and the cellular load. In particular, in our simulation, the total transmission cost can be reduced by 42.5%-55.2% and the cellular load can be reduced by 21.5%-56.4%.
We present a wave generalization of the classic Schwarzschild method for constructing self-consistent halos -- such a halo consists of a suitable superposition of waves instead of particle orbits, chosen to yield a desired mean density profile. As an
illustration, the method is applied to spherically symmetric halos. We derive an analytic relation between the particle distribution function and the wave superposition amplitudes, and show how it simplifies in the high energy (WKB) limit. We verify the stability of such constructed halos by numerically evolving the Schrodinger-Poisson system. The algorithm provides an efficient and accurate way to simulate the time-dependent halo substructures from wave interference. We use this method to construct halos with a variety of density profiles, all of which have a core from the ground-state wave function, though the core-halo relation need not be the standard one.
Event detection has long been troubled by the emph{trigger curse}: overfitting the trigger will harm the generalization ability while underfitting it will hurt the detection performance. This problem is even more severe in few-shot scenario. In this
paper, we identify and solve the trigger curse problem in few-shot event detection (FSED) from a causal view. By formulating FSED with a structural causal model (SCM), we found that the trigger is a confounder of the context and the result, which makes previous FSED methods much easier to overfit triggers. To resolve this problem, we propose to intervene on the context via backdoor adjustment during training. Experiments show that our method significantly improves the FSED on ACE05, MAVEN and KBP17 datasets.
Conventional entity typing approaches are based on independent classification paradigms, which make them difficult to recognize inter-dependent, long-tailed and fine-grained entity types. In this paper, we argue that the implicitly entailed extrinsic
and intrinsic dependencies between labels can provide critical knowledge to tackle the above challenges. To this end, we propose emph{Label Reasoning Network(LRN)}, which sequentially reasons fine-grained entity labels by discovering and exploiting label dependencies knowledge entailed in the data. Specifically, LRN utilizes an auto-regressive network to conduct deductive reasoning and a bipartite attribute graph to conduct inductive reasoning between labels, which can effectively model, learn and reason complex label dependencies in a sequence-to-set, end-to-end manner. Experiments show that LRN achieves the state-of-the-art performance on standard ultra fine-grained entity typing benchmarks, and can also resolve the long tail label problem effectively.
Navigating a large-scaled robot in unknown and cluttered height-constrained environments is challenging. Not only is a fast and reliable planning algorithm required to go around obstacles, the robot should also be able to change its intrinsic dimensi
on by crouching in order to travel underneath height constrained regions. There are few mobile robots that are capable of handling such a challenge, and bipedal robots provide a solution. However, as bipedal robots have nonlinear and hybrid dynamics, trajectory planning while ensuring dynamic feasibility and safety on these robots is challenging. This paper presents an end-to-end vision-aided autonomous navigation framework which leverages three layers of planners and a variable walking height controller to enable bipedal robots to safely explore height-constrained environments. A vertically actuated Spring-Loaded Inverted Pendulum (vSLIP) model is introduced to capture the robot coupled dynamics of planar walking and vertical walking height. This reduced-order model is utilized to optimize for long-term and short-term safe trajectory plans. A variable walking height controller is leveraged to enable the bipedal robot to maintain stable periodic walking gaits while following the planned trajectory. The entire framework is tested and experimentally validated using a bipedal robot Cassie. This demonstrates reliable autonomy to drive the robot to safely avoid obstacles while walking to the goal location in various kinds of height-constrained cluttered environments.
Rapid developments of quantum information technology show promising opportunities for simulating quantum field theory in near-term quantum devices. In this work, we formulate the theory of (time-dependent) variational quantum simulation, explicitly d
esigned for quantum simulation of quantum field theory. We develop hybrid quantum-classical algorithms for crucial ingredients in particle scattering experiments, including encoding, state preparation, and time evolution, with several numerical simulations to demonstrate our algorithms in the 1+1 dimensional $lambda phi^4$ quantum field theory. These algorithms could be understood as near-term analogs of the Jordan-Lee-Preskill algorithm, the basic algorithm for simulating quantum field theory using universal quantum devices. Our contribution also includes a bosonic version of the Unitary Coupled Cluster ansatz with physical interpretation in quantum field theory, a discussion about the subspace fidelity, a comparison among different bases in the 1+1 dimensional $lambda phi^4$ theory, and the spectral crowding in the quantum field theory simulation.
Direct automatic segmentation of objects from 3D medical imaging, such as magnetic resonance (MR) imaging, is challenging as it often involves accurately identifying a number of individual objects with complex geometries within a large volume under i
nvestigation. To address these challenges, most deep learning approaches typically enhance their learning capability by substantially increasing the complexity or the number of trainable parameters within their models. Consequently, these models generally require long inference time on standard workstations operating clinical MR systems and are restricted to high-performance computing hardware due to their large memory requirement. Further, to fit 3D dataset through these large models using limited computer memory, trade-off techniques such as patch-wise training are often used which sacrifice the fine-scale geometric information from input images which could be clinically significant for diagnostic purposes. To address these challenges, we present a compact convolutional neural network with a shallow memory footprint to efficiently reduce the number of model parameters required for state-of-art performance. This is critical for practical employment as most clinical environments only have low-end hardware with limited computing power and memory. The proposed network can maintain data integrity by directly processing large full-size 3D input volumes with no patches required and significantly reduces the computational time required for both training and inference. We also propose a novel loss function with extra shape constraint to improve the accuracy for imbalanced classes in 3D MR images.
