arXiv subjects

Qiang Ye

Publications and source records attributed to Qiang Ye.

At least 37 records · Page 2Linked to original sources

arXiv2023

Accelerated Sparse Recovery via Gradient Descent with Nonlinear Conjugate Gradient Momentum

This paper applies an idea of adaptive momentum for the nonlinear conjugate gradient to accelerate optimization problems in sparse recovery. Specifically, we consider two types of minimization problems: a (single) differentiable function and the sum of a non-smooth function and a differentiable function. In the first case, we adopt a fixed step size to avoid the traditional line search and establish the convergence analysis of the proposed algorithm for a quadratic problem. This acceleration is further incorporated with an operator splitting technique to deal with the non-smooth function in the second case. We use the convex $\ell_1$ and the nonconvex $\ell_1-\ell_2$ functionals as two case studies to demonstrate the efficiency of the proposed approaches over traditional methods.

math.OC↗

arXiv2022

SCP-GAN: Self-Correcting Discriminator Optimization for Training Consistency Preserving Metric GAN on Speech Enhancement Tasks

In recent years, Generative Adversarial Networks (GANs) have produced significantly improved results in speech enhancement (SE) tasks. They are difficult to train, however. In this work, we introduce several improvements to the GAN training schemes, which can be applied to most GAN-based SE models. We propose using consistency loss functions, which target the inconsistency in time and time-frequency domains caused by Fourier and Inverse Fourier Transforms. We also present self-correcting optimization for training a GAN discriminator on SE tasks, which helps avoid "harmful" training directions for parts of the discriminator loss function. We have tested our proposed methods on several state-of-the-art GAN-based SE models and obtained consistent improvements, including new state-of-the-art results for the Voice Bank+DEMAND dataset.

cs.SD↗

arXiv2022

Breaking Time Invariance: Assorted-Time Normalization for RNNs

Methods such as Layer Normalization (LN) and Batch Normalization (BN) have proven to be effective in improving the training of Recurrent Neural Networks (RNNs). However, existing methods normalize using only the instantaneous information at one particular time step, and the result of the normalization is a preactivation state with a time-independent distribution. This implementation fails to account for certain temporal differences inherent in the inputs and the architecture of RNNs. Since these networks share weights across time steps, it may also be desirable to account for the connections between time steps in the normalization scheme. In this paper, we propose a normalization method called Assorted-Time Normalization (ATN), which preserves information from multiple consecutive time steps and normalizes using them. This setup allows us to introduce longer time dependencies into the traditional normalization methods without introducing any new trainable parameters. We present theoretical derivations for the gradient propagation and prove the weight scaling invariance property. Our experiments applying ATN to LN demonstrate consistent improvement on various tasks, such as Adding, Copying, and Denoise Problems and Language Modeling Problems.

cs.LG↗

arXiv2022

Orthogonal Gated Recurrent Unit with Neumann-Cayley Transformation

In recent years, using orthogonal matrices has been shown to be a promising approach in improving Recurrent Neural Networks (RNNs) with training, stability, and convergence, particularly, to control gradients. While Gated Recurrent Unit (GRU) and Long Short Term Memory (LSTM) architectures address the vanishing gradient problem by using a variety of gates and memory cells, they are still prone to the exploding gradient problem. In this work, we analyze the gradients in GRU and propose the usage of orthogonal matrices to prevent exploding gradient problems and enhance long-term memory. We study where to use orthogonal matrices and we propose a Neumann series-based Scaled Cayley transformation for training orthogonal matrices in GRU, which we call Neumann-Cayley Orthogonal GRU, or simply NC-GRU. We present detailed experiments of our model on several synthetic and real-world tasks, which show that NC-GRU significantly outperforms GRU as well as several other RNNs.

cs.LG↗

arXiv2022

AUTM Flow: Atomic Unrestricted Time Machine for Monotonic Normalizing Flows

Nonlinear monotone transformations are used extensively in normalizing flows to construct invertible triangular mappings from simple distributions to complex ones. In existing literature, monotonicity is usually enforced by restricting function classes or model parameters and the inverse transformation is often approximated by root-finding algorithms as a closed-form inverse is unavailable. In this paper, we introduce a new integral-based approach termed "Atomic Unrestricted Time Machine (AUTM)", equipped with unrestricted integrands and easy-to-compute explicit inverse. AUTM offers a versatile and efficient way to the design of normalizing flows with explicit inverse and unrestricted function classes or parameters. Theoretically, we present a constructive proof that AUTM is universal: all monotonic normalizing flows can be viewed as limits of AUTM flows. We provide a concrete example to show how to approximate any given monotonic normalizing flow using AUTM flows with guaranteed convergence. The result implies that AUTM can be used to transform an existing flow into a new one equipped with explicit inverse and unrestricted parameters. The performance of the new approach is evaluated on high dimensional density estimation, variational inference and image generation. Experiments demonstrate superior speed and memory efficiency of AUTM.

cs.LG↗

arXiv2022

Symmetry Structured Convolutional Neural Networks

We consider Convolutional Neural Networks (CNNs) with 2D structured features that are symmetric in the spatial dimensions. Such networks arise in modeling pairwise relationships for a sequential recommendation problem, as well as secondary structure inference problems of RNA and protein sequences. We develop a CNN architecture that generates and preserves the symmetry structure in the network's convolutional layers. We present parameterizations for the convolutional kernels that produce update rules to maintain symmetry throughout the training. We apply this architecture to the sequential recommendation problem, the RNA secondary structure inference problem, and the protein contact map prediction problem, showing that the symmetric structured networks produce improved results using fewer numbers of machine parameters.

stat.ML↗

arXiv2022

Batch Normalization Preconditioning for Neural Network Training

Batch normalization (BN) is a popular and ubiquitous method in deep learning that has been shown to decrease training time and improve generalization performance of neural networks. Despite its success, BN is not theoretically well understood. It is not suitable for use with very small mini-batch sizes or online learning. In this paper, we propose a new method called Batch Normalization Preconditioning (BNP). Instead of applying normalization explicitly through a batch normalization layer as is done in BN, BNP applies normalization by conditioning the parameter gradients directly during training. This is designed to improve the Hessian matrix of the loss function and hence convergence during training. One benefit is that BNP is not constrained on the mini-batch size and works in the online learning setting. Furthermore, its connection to BN provides theoretical insights on how BN improves training and how BN is applied to special architectures such as convolutional neural networks. For a theoretical foundation, we also present a novel Hessian condition number based convergence theory for a locally convex but not strong-convex loss, which is applicable to networks with a scale-invariant property.

cs.LG↗

arXiv2021

Symmetry of magnetic correlations in spin-triplet superconductor UTe2

The temperature dependence of the low-energy magnetic excitations in the spin-triplet superconductor UTe$_2$ was measured via inelastic neutron scattering in the normal and superconducting states. The imaginary part of the dynamic susceptibility follows the behavior of interband correlations in a hybridized Kondo lattice with an appropriate characteristic energy. These excitations are a lower-dimensional analogue of phenomena observed in other Kondo lattice materials, such that their presence is not necessarily due to dominance of ferromagnetic or antiferromagnetic correlations. The onset of superconductivity alters the magnetic excitations noticeably on the same energy scales, suggesting that these changes originate from additional electronic structure modification.

cond-mat.str-el↗

arXiv2021

Two-stage short-term wind power forecasting algorithm using different feature-learning models

Two-stage ensemble-based forecasting methods have been studied extensively in the wind power forecasting field. However, deep learning-based wind power forecasting studies have not investigated two aspects. In the first stage, different learning structures considering multiple inputs and multiple outputs have not been discussed. In the second stage, the model extrapolation issue has not been investigated. Therefore, we develop four deep neural networks for the first stage to learn data features considering the input-and-output structure. We then explore the model extrapolation issue in the second stage using different modeling methods. Considering the overfitting issue, we propose a new moving window-based algorithm using a validation set in the first stage to update the training data in both stages with two different moving window processes.Experiments were conducted at three wind farms, and the results demonstrate that the model with single input multiple output structure obtains better forecasting accuracy compared to existing models. In addition, the ridge regression method results in a better ensemble model that can further improve forecasting accuracy compared to existing machine learning methods. Finally, the proposed two-stage forecasting algorithm can generate more accurate and stable results than existing algorithms.

eess.SP↗

arXiv2021

Adaptive Weighted Discriminator for Training Generative Adversarial Networks

Generative adversarial network (GAN) has become one of the most important neural network models for classical unsupervised machine learning. A variety of discriminator loss functions have been developed to train GAN's discriminators and they all have a common structure: a sum of real and fake losses that only depends on the actual and generated data respectively. One challenge associated with an equally weighted sum of two losses is that the training may benefit one loss but harm the other, which we show causes instability and mode collapse. In this paper, we introduce a new family of discriminator loss functions that adopts a weighted sum of real and fake parts, which we call adaptive weighted loss functions or aw-loss functions. Using the gradients of the real and fake parts of the loss, we can adaptively choose weights to train a discriminator in the direction that benefits the GAN's stability. Our method can be potentially applied to any discriminator model with a loss that is a sum of the real and fake parts. Experiments validated the effectiveness of our loss functions on an unconditional image generation task, improving the baseline results by a significant margin on CIFAR-10, STL-10, and CIFAR-100 datasets in Inception Scores and FID.

cs.LG↗

arXiv2020

Stochastic Gradient Descent with Nonlinear Conjugate Gradient-Style Adaptive Momentum

Momentum plays a crucial role in stochastic gradient-based optimization algorithms for accelerating or improving training deep neural networks (DNNs). In deep learning practice, the momentum is usually weighted by a well-calibrated constant. However, tuning hyperparameters for momentum can be a significant computational burden. In this paper, we propose a novel \emph{adaptive momentum} for improving DNNs training; this adaptive momentum, with no momentum related hyperparameter required, is motivated by the nonlinear conjugate gradient (NCG) method. Stochastic gradient descent (SGD) with this new adaptive momentum eliminates the need for the momentum hyperparameter calibration, allows a significantly larger learning rate, accelerates DNN training, and improves final accuracy and robustness of the trained DNNs. For instance, SGD with this adaptive momentum reduces classification errors for training ResNet110 for CIFAR10 and CIFAR100 from $5.25\%$ to $4.64\%$ and $23.75\%$ to $20.03\%$, respectively. Furthermore, SGD with the new adaptive momentum also benefits adversarial training and improves adversarial robustness of the trained DNNs.

cs.LG↗

arXiv2020

Secure Non-Orthogonal Multiple Access: An Interference Engineering Perspective

Non-orthogonal multiple access (NOMA) is an efficient approach that can improve spectrum utilization and support massive connectivity for next-generation wireless networks. However, over a wireless channel, the superimposed NOMA signals are highly susceptible to eavesdropping, potentially leading to severe leakage of confidential information. In this article, we unleash the potential of network interference and exploit it constructively to enhance physical-layer security in NOMA networks. Particularly, three different types of network interference, including artificial noise, specifically-designed jamming signals, and inter-user interference, are well engineered to intentionally reduce information leakage while mitigating the effect on signal reception quality of legitimate users, thereby significantly enhancing the transmission security of NOMA. Furthermore, we propose interference engineering strategies for more advanced full-duplex NOMA, intelligent reflecting surface NOMA, cognitive radio NOMA, and multi-cell NOMA networks, and discuss several open research problems and challenges, which could inspire innovative interference engineering designs for secure NOMA communications.

cs.IT↗

arXiv2020

Improving RNA secondary structure prediction via state inference with deep recurrent neural networks

The problem of determining which nucleotides of an RNA sequence are paired or unpaired in the secondary structure of an RNA, which we call RNA state inference, can be studied by different machine learning techniques. Successful state inference of RNA sequences can be used to generate auxiliary information for data-directed RNA secondary structure prediction. Bidirectional long short-term memory (LSTM) neural networks have emerged as a powerful tool that can model global nonlinear sequence dependencies and have achieved state-of-the-art performances on many different classification problems. This paper presents a practical approach to RNA secondary structure inference centered around a deep learning method for state inference. State predictions from a deep bidirectional LSTM are used to generate synthetic SHAPE data that can be incorporated into RNA secondary structure prediction via the Nearest Neighbor Thermodynamic Model (NNTM). This method produces predicted secondary structures for a diverse test set of 16S ribosomal RNA that are, on average, 25 percentage points more accurate than undirected MFE structures. These improvements range from several percentage points for some sequences to nearly 50 percentage points for others. Accuracy is highly dependent on the success of our state inference method, and investigating the global features of our state predictions reveals that accuracy of both our state inference and structure inference methods are highly dependent on the similarity of the sequence to the dataset. This paper presents a deep learning state inference tool, trained and tested on 16S ribosomal RNA. Converting these state predictions into synthetic SHAPE data with which to direct NNTM can result in large improvements in secondary structure prediction accuracy, as shown on a test set of 16S rRNA.

q-bio.BM↗

arXiv2020

A Virtual Network Customization Framework for Multicast Services in NFV-enabled Core Networks

The paradigm of network function virtualization (NFV) with the support of software defined networking (SDN) emerges as a promising approach for customizing network services in fifth generation (5G) networks. In this paper, a multicast service orchestration framework is presented, where joint traffic routing and virtual network function (NF) placement are studied for accommodating multicast services over an NFV-enabled physical substrate network. First, we investigate a joint routing and NF placement problem for a single multicast request accommodated over a physical substrate network, with both single-path and multipath traffic routing. The joint problem is formulated as a mixed integer linear programming (MILP) problem to minimize the function and link provisioning costs, under the physical network resource constraints, flow conservation constraints, and NF placement rules; Second, we develop an MILP formulation that jointly handles the static embedding of multiple service requests over the physical substrate network, where we determine the optimal combination of multiple services for embedding and their joint routing and placement configurations, such that the aggregate throughput of the physical substrate is maximized, while the function and link provisioning costs are minimized. Since the presented problem formulations are NP-hard, low complexity heuristic algorithms are proposed to find an efficient solution for both single-path and multipath routing scenarios. Simulation results are presented to demonstrate the effectiveness and accuracy of the proposed heuristic algorithms.

cs.NI↗

arXiv2019

Eigenvalue Normalized Recurrent Neural Networks for Short Term Memory

Several variants of recurrent neural networks (RNNs) with orthogonal or unitary recurrent matrices have recently been developed to mitigate the vanishing/exploding gradient problem and to model long-term dependencies of sequences. However, with the eigenvalues of the recurrent matrix on the unit circle, the recurrent state retains all input information which may unnecessarily consume model capacity. In this paper, we address this issue by proposing an architecture that expands upon an orthogonal/unitary RNN with a state that is generated by a recurrent matrix with eigenvalues in the unit disc. Any input to this state dissipates in time and is replaced with new inputs, simulating short-term memory. A gradient descent algorithm is derived for learning such a recurrent matrix. The resulting method, called the Eigenvalue Normalized RNN (ENRNN), is shown to be highly competitive in several experiments.

cs.LG↗

arXiv2019

Optimal Dynamic Multi-Resource Management in Earth Observation Oriented Space Information Networks

Space information network (SIN) is an innovative networking architecture to achieve near-real-time mass data observation, processing and transmission over the globe. In the SIN environment, it is essential to coordinate multi-dimensional heterogeneous resources (i.e., observation resource, computation resource and transmission resource) to improve network performance. However, the time varying property of both the observation resource and transmission resource is not fully exploited in existing studies. Dynamic resource management according to instantaneous channel conditions has a potential to enhance network performance. To this end, in this paper, we study the multi-resource dynamic management problem, considering stochastic observation and transmission channel conditions in SINs. Specifically, we develop an aggregate optimization framework for observation scheduling, compression ratio selection and transmission scheduling, and formulate a flow optimization problem based on extended time expanded graph (ETEG) to maximize the sum network utility. Then, we equivalently transform the flow optimization problem on ETEG as a queue stability-related stochastic optimization problem. An online algorithm is proposed to solve the problem in a slot-by-slot manner by exploiting the Lyapunov optimization technique. Performance analysis shows that the proposed algorithm achieves close-to-optimal network utility while guaranteeing bounded queue occupancy. Extensive simulation results further validate the efficiency of the proposed algorithm and evaluate the impacts of various network parameters on the algorithm performance.

eess.SY↗

arXiv2019

On regularization for a convolutional kernel in neural networks

Convolutional neural network is an important model in deep learning. To avoid exploding/vanishing gradient problems and to improve the generalizability of a neural network, it is desirable to have a convolution operation that nearly preserves the norm, or to have the singular values of the transformation matrix corresponding to a convolutional kernel bounded around $1$. We propose a penalty function that can be used in the optimization of a convolutional neural network to constrain the singular values of the transformation matrix around $1$. We derive an algorithm to carry out the gradient descent minimization of this penalty function in terms of convolution kernels. Numerical examples are presented to demonstrate the effectiveness of the method.

cs.LG↗

arXiv2019

Delay-Aware Flow Migration for Embedded Services in 5G Core Networks

Service-oriented virtual network deployment is based on statistical resource demands of different services, while data traffic from each service fluctuates over time. In this paper, a delay-aware flow migration problem for embedded services is studied to meet end-to-end (E2E) delay requirement with time-varying traffic. A non-convex multi-objective mixed integer optimization problem is formulated, addressing the trade-off between maximum load balancing and minimum reconfiguration overhead due to flow migrations, under processing and transmission resource constraints and QoS requirement constraints. Since the original problem is non-solvable in optimization solvers due to unsupported types of quadratic constraints, it is transformed to a tractable mixed integer quadratically constrained programming (MIQCP) problem. The optimality gap between the two problems is proved to be zero, so we can obtain the optimum of the original problem through solving the MIQCP problem with some post-processing. Numerical results are presented to demonstrate the aforementioned trade-off, as well as the benefit from flow migration in terms of E2E delay performance guarantee.

cs.NI↗