arXiv subjects

Liang Huang

Publications and source records attributed to Liang Huang.

At least 55 records · Page 3Linked to original sources

arXiv2022

Universal cover-time distribution of heterogeneous random walks

The cover-time problem, i.e., time to visit every site in a system, is one of the key issues of random walks with wide applications in natural, social, and engineered systems. Addressing the full distribution of cover times for random walk on complex structures has been a long-standing challenge and has attracted persistent efforts. Yet, the known results are essentially limited to homogeneous systems, where different sites are on an equal footing and have identical or close mean first-passage times, such as random walks on a torus. In contrast, realistic random walks are prevailingly heterogeneous with diversified mean first-passage times. Does a universal distribution still exist? Here, by considering the most general situations, we uncover a generalized rescaling relation for the cover time, exploiting the diversified mean first-passage times that have not been accounted for before. This allows us to concretely establish a universal distribution of the rescaled cover times for heterogeneous random walks, which turns out to be the Gumbel universality class that is ubiquitous for a large family of extreme value statistics. Our analysis is based on the transfer matrix framework, which is generic that besides heterogeneity, it is also robust against biased protocols, directed links, and self-connecting loops. The finding is corroborated with extensive numerical simulations of diverse heterogeneous non-compact random walks on both model and realistic topological structures. Our new technical ingredient may be exploited for other extreme value or ergodicity problems with nonidentical distributions.

cond-mat.stat-mech↗

arXiv2022

Experimental demonstration of genuine tripartite nonlocality under strict locality conditions

Nonlocality captures one of the counterintuitive features of nature that defies classical intuition. Recent investigations reveal that our physical world's nonlocality is at least tripartite; i.e., genuinely tripartite nonlocal correlations in nature cannot be reproduced by any causal theory involving bipartite nonclassical resources and unlimited shared randomness. Here, by allowing the fair sampling assumption and postselection, we experimentally demonstrate such genuine tripartite nonlocality in a network under strict locality constraints that are ensured by spacelike separating all relevant events and employing fast quantum random number generators and high-speed polarization measurements. In particular, for a photonic quantum triangular network we observe a locality-loophole-free violation of the Bell-type inequality by 7.57 standard deviations for a postselected tripartite Greenberger-Horne-Zeilinger state of fidelity $(93.13 \pm 0.24)\%$, which convincingly disproves the possibility of simulating genuine tripartite nonlocality by bipartite nonlocal resources with globally shared randomness.

quant-ph↗

arXiv2022

A$^3$T: Alignment-Aware Acoustic and Text Pretraining for Speech Synthesis and Editing

Recently, speech representation learning has improved many speech-related tasks such as speech recognition, speech classification, and speech-to-text translation. However, all the above tasks are in the direction of speech understanding, but for the inverse direction, speech synthesis, the potential of representation learning is yet to be realized, due to the challenging nature of generating high-quality speech. To address this problem, we propose our framework, Alignment-Aware Acoustic-Text Pretraining (A$^3$T), which reconstructs masked acoustic signals with text input and acoustic-text alignment during training. In this way, the pretrained model can generate high quality reconstructed spectrogram, which can be applied to the speech editing and unseen speaker TTS directly. Experiments show A$^3$T outperforms SOTA models on speech editing, and improves multi-speaker speech synthesis without the external speaker verification model.

eess.AS↗

arXiv2022

PaddleSpeech: An Easy-to-Use All-in-One Speech Toolkit

PaddleSpeech is an open-source all-in-one speech toolkit. It aims at facilitating the development and research of speech processing technologies by providing an easy-to-use command-line interface and a simple code structure. This paper describes the design philosophy and core architecture of PaddleSpeech to support several essential speech-to-text and text-to-speech tasks. PaddleSpeech achieves competitive or state-of-the-art performance on various speech datasets and implements the most popular methods. It also provides recipes and pretrained models to quickly reproduce the experimental results in this paper. PaddleSpeech is publicly avaiable at https://github.com/PaddlePaddle/PaddleSpeech.

eess.AS↗

arXiv2022

A Fast Attention Network for Joint Intent Detection and Slot Filling on Edge Devices

Intent detection and slot filling are two main tasks in natural language understanding and play an essential role in task-oriented dialogue systems. The joint learning of both tasks can improve inference accuracy and is popular in recent works. However, most joint models ignore the inference latency and cannot meet the need to deploy dialogue systems at the edge. In this paper, we propose a Fast Attention Network (FAN) for joint intent detection and slot filling tasks, guaranteeing both accuracy and latency. Specifically, we introduce a clean and parameter-refined attention module to enhance the information exchange between intent and slot, improving semantic accuracy by more than 2%. FAN can be implemented on different encoders and delivers more accurate models at every speed level. Our experiments on the Jetson Nano platform show that FAN inferences fifteen utterances per second with a small accuracy drop, showing its effectiveness and efficiency on edge devices.

cs.CL↗

arXiv2022

Data-Driven Adaptive Simultaneous Machine Translation

In simultaneous translation (SimulMT), the most widely used strategy is the wait-k policy thanks to its simplicity and effectiveness in balancing translation quality and latency. However, wait-k suffers from two major limitations: (a) it is a fixed policy that can not adaptively adjust latency given context, and (b) its training is much slower than full-sentence translation. To alleviate these issues, we propose a novel and efficient training scheme for adaptive SimulMT by augmenting the training corpus with adaptive prefix-to-prefix pairs, while the training complexity remains the same as that of training full-sentence translation models. Experiments on two language pairs show that our method outperforms all strong baselines in terms of translation quality and latency.

cs.CL↗

arXiv2022

Prediction and Control of Focal Seizure Spread: Random Walk with Restart on Heterogeneous Brain Networks

Whole-brain models offer a promising method of predicting seizure spread, which is critical for successful surgery treatment of focal epilepsy. Existing methods are largely based on structural connectome, which ignores the effects of heterogeneity in regional excitability of brains. In this study, we used a whole-brain model to show that heterogeneity in nodal excitability had a significant impact on seizure propagation in the networks, and compromised the prediction accuracy with structural connections. We then addressed this problem with an algorithm based on random walk with restart on graphs. We demonstrated that by establishing a relationship between the restarting probability and the excitability for each node, this algorithm could significantly improve the seizure spread prediction accuracy in heterogeneous networks, and was more robust against the extent of heterogeneity. We also strategized surgical seizure control as a process to identify and remove the key nodes (connections) responsible for the early spread of seizures from the focal region. Compared to strategies based on structural connections, virtual surgery with a strategy based on mRWER generated outcomes with a high success rate while maintaining low damage to the brain by removing fewer anatomical connections. These findings may have potential applications in developing personalized surgery strategies for epilepsy.

q-bio.NC↗

arXiv2022

Algorithm for Optimized mRNA Design Improves Stability and Immunogenicity

Messenger RNA (mRNA) vaccines are being used for COVID-19, but still suffer from the critical issue of mRNA instability and degradation, which is a major obstacle in the storage, distribution, and efficacy of the vaccine. Previous work showed that optimizing secondary structure stability lengthens mRNA half-life, which, together with optimal codons, increases protein expression. Therefore, a principled mRNA design algorithm must optimize both structural stability and codon usage to improve mRNA efficiency. However, due to synonymous codons, the mRNA design space is prohibitively large, e.g., there are $\sim\!10^{632}$ mRNAs for the SARS-CoV-2 Spike protein, which poses insurmountable challenges to previous methods. Here we provide a surprisingly simple solution to this hard problem by reducing it to a classical problem in computational linguistics, where finding the optimal mRNA is akin to finding the most likely sentence among similar sounding alternatives. Our algorithm, named LinearDesign, takes only 11 minutes for the Spike protein, and can jointly optimize stability and codon usage. Experimentally, without chemical modification, our designs substantially improve mRNA half-life and protein expression in vitro, and dramatically increase antibody response by up to 23$\times$ in vivo, compared to the codon-optimized benchmark. Our work enables the exploration of highly stable and efficient designs that are previously unreachable and is a timely tool not only for vaccines but also for mRNA medicine encoding all therapeutic proteins (e.g., monoclonal antibodies and anti-cancer drugs).

q-bio.BM↗

arXiv2022

Experimental refutation of real-valued quantum mechanics under strict locality conditions

Physicists describe nature using mathematics as the natural language, and for quantum mechanics, it prefers to use complex numbers. However, whether complex numbers are really necessary for the theory has been debated ever since its birth. Recently, it has been shown that a three-party correlation created in entanglement swapping scenarios comprising independent states and measurements cannot be reproduced using only real numbers. Previous experiments have conceptually supported the predication, yet not satisfying the independent state preparations and measurements simultaneously. Here, we implement such a test with two truly independent sources delivering entangled photons to three parties under strict locality conditions. By employing fast quantum random number generators and high-speed polarization measurements, we space-like separate all relevant events to ensure independent state preparations and measurements, and close locality loopholes simultaneously. Our results violate the real number bound of 7.66 by 5.30 standard deviations, hence rejecting the universal validity of the real-valued quantum mechanics to describe nature.

quant-ph↗

arXiv2021

Optimal weak estimates for Riesz potentials

In this note we prove a sharp reverse weak estimate for Riesz potentials $$\|I_{s}(f)\|_{L^{\frac{n}{n-s},\infty}}\geq γ_sv_{n}^{\frac{n-s}{n}}\|f\|_{L^1}~~\text{for}~~0<f\in {L^1(\mathbb{R}^n)},$$ where $γ_s=2^{-s}π^{-\frac{n}{2}}\frac{Γ(\frac{n-s}{2})}{Γ(\frac{s}{2})}$. We also consider the behavior of the best constant $\mathcal{C}_{n,s}$ of weak type estimate for Riesz potentials, and we prove $\mathcal{C}_{n,s}=O(\frac{γ_s}{s})$ as $s\rightarrow 0$.

math.CA↗

arXiv2021

SigNet: A Novel Deep Learning Framework for Radio Signal Classification

Deep learning methods achieve great success in many areas due to their powerful feature extraction capabilities and end-to-end training mechanism, and recently they are also introduced for radio signal modulation classification. In this paper, we propose a novel deep learning framework called SigNet, where a signal-to-matrix (S2M) operator is adopted to convert the original signal into a square matrix first and is co-trained with a follow-up CNN architecture for classification. This model is further accelerated by integrating 1D convolution operators, leading to the upgraded model SigNet2.0. The simulations on two signal datasets show that both SigNet and SigNet2.0 outperform a number of well-known baselines. More interestingly, our proposed models behave extremely well in small-sample learning when only a small training dataset is provided. They can achieve a relatively high accuracy even when 1\% training data are kept, while other baseline models may lose their effectiveness much more quickly as the datasets get smaller. Such result suggests that SigNet/SigNet2.0 could be extremely useful in the situations where labeled signal data are difficult to obtain. The visualization of the output features of our models demonstrates that our model can well divide different modulation types of signals in the feature hyper-space.

eess.SP↗

arXiv2021

An Integrated Optimization-Learning Framework for Online Combinatorial Computation Offloading in MEC Networks

Mobile edge computing (MEC) is a promising paradigm to accommodate the increasingly prosperous delay-sensitive and computation-intensive applications in 5G systems. To achieve optimum computation performance in a dynamic MEC environment, mobile devices often need to make online decisions on whether to offload the computation tasks to nearby edge terminals under the uncertainty of future system information (e.g., random wireless channel gain and task arrivals). The design of an efficient online offloading algorithm is challenging. On one hand, the fast-varying edge environment requires frequently solving a hard combinatorial optimization problem where the integer offloading decision and continuous resource allocation variables are strongly coupled. On the other hand, the uncertainty of future system states makes it hard for the online decisions to satisfy long-term system constraints. To address these challenges, this article overviews the existing methods and introduces a novel framework that efficiently integrates model-based optimization and model-free learning techniques. Besides, we suggest some promising future research directions of online computation offloading control in MEC networks.

cs.IT↗

arXiv2021

Improving Simultaneous Translation by Incorporating Pseudo-References with Fewer Reorderings

Simultaneous translation is vastly different from full-sentence translation, in the sense that it starts translation before the source sentence ends, with only a few words delay. However, due to the lack of large-scale, high-quality simultaneous translation datasets, most such systems are still trained on conventional full-sentence bitexts. This is far from ideal for the simultaneous scenario due to the abundance of unnecessary long-distance reorderings in those bitexts. We propose a novel method that rewrites the target side of existing full-sentence corpora into simultaneous-style translation. Experiments on Zh->En and Ja->En simultaneous translation show substantial improvements (up to +2.7 BLEU) with the addition of these generated pseudo-references.

cs.CL↗

arXiv2021

Fused Acoustic and Text Encoding for Multimodal Bilingual Pretraining and Speech Translation

Recently, representation learning for text and speech has successfully improved many language related tasks. However, all existing methods suffer from two limitations: (a) they only learn from one input modality, while a unified representation for both speech and text is needed by tasks such as end-to-end speech translation, and as a result,(b) they can not exploit various large-scale text and speech data and their performance is limited by the scarcity of parallel speech translation data.To address these problems, we propose a Fused Acoustic and Text Masked Language Model (FAT-MLM) which jointly learns a unified representation for both acoustic and text input from various types of corpora including parallel data for speech recognition and machine translation, and even pure speech and text data. Within this cross-modal representation learning framework, we further present an end-to-end model for Fused Acoustic and Text Speech Translation (FAT-ST). Experiments on three translation directions show that by fine-tuning from FAT-MLM, our proposed speech translation models substantially improve translation quality by up to +5.9 BLEU.

cs.CL↗

arXiv2021

Computation Rate Maximum for Mobile Terminals in UAV-assisted Wireless Powered MEC Networks with Fairness Constraint

This paper investigates an unmanned aerial vehicle (UAV)-assisted wireless powered mobile-edge computing (MEC) system, where the UAV powers the mobile terminals by wireless power transfer (WPT) and provides computation service for them. We aim to maximize the computation rate of terminals while ensuring fairness among them. Considering the random trajectories of mobile terminals, we propose a soft actor-critic (SAC)-based UAV trajectory planning and resource allocation (SAC-TR) algorithm, which combines off-policy and maximum entropy reinforcement learning to promote the convergence of the algorithm. We design the reward as a heterogeneous function of computation rate, fairness, and reaching of destination. Simulation results show that SAC-TR can quickly adapt to varying network environments and outperform representative benchmarks in a variety of situations.

cs.NI↗

arXiv2021

Decoupling recognition and transcription in Mandarin ASR

Much of the recent literature on automatic speech recognition (ASR) is taking an end-to-end approach. Unlike English where the writing system is closely related to sound, Chinese characters (Hanzi) represent meaning, not sound. We propose factoring audio -> Hanzi into two sub-tasks: (1) audio -> Pinyin and (2) Pinyin -> Hanzi, where Pinyin is a system of phonetic transcription of standard Chinese. Factoring the audio -> Hanzi task in this way achieves 3.9% CER (character error rate) on the Aishell-1 corpus, the best result reported on this dataset so far.

cs.CL↗

arXiv2021

The Role of Phonetic Units in Speech Emotion Recognition

We propose a method for emotion recognition through emotiondependent speech recognition using Wav2vec 2.0. Our method achieved a significant improvement over most previously reported results on IEMOCAP, a benchmark emotion dataset. Different types of phonetic units are employed and compared in terms of accuracy and robustness of emotion recognition within and across datasets and languages. Models of phonemes, broad phonetic classes, and syllables all significantly outperform the utterance model, demonstrating that phonetic units are helpful and should be incorporated in speech emotion recognition. The best performance is from using broad phonetic classes. Further research is needed to investigate the optimal set of broad phonetic classes for the task of emotion recognition. Finally, we found that Wav2vec 2.0 can be fine-tuned to recognize coarser-grained or larger phonetic units than phonemes, such as broad phonetic classes and syllables.

cs.CL↗

arXiv2021

Structure-aware Interactive Graph Neural Networks for the Prediction of Protein-Ligand Binding Affinity

Drug discovery often relies on the successful prediction of protein-ligand binding affinity. Recent advances have shown great promise in applying graph neural networks (GNNs) for better affinity prediction by learning the representations of protein-ligand complexes. However, existing solutions usually treat protein-ligand complexes as topological graph data, thus the biomolecular structural information is not fully utilized. The essential long-range interactions among atoms are also neglected in GNN models. To this end, we propose a structure-aware interactive graph neural network (SIGN) which consists of two components: polar-inspired graph attention layers (PGAL) and pairwise interactive pooling (PiPool). Specifically, PGAL iteratively performs the node-edge aggregation process to update embeddings of nodes and edges while preserving the distance and angle information among atoms. Then, PiPool is adopted to gather interactive edges with a subsequent reconstruction loss to reflect the global interactions. Exhaustive experimental study on two benchmarks verifies the superiority of SIGN.

q-bio.QM↗