BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Jun Du

Jun Du contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
eess.ASSoundComputer VisionMachine LearningArtificial Intelligenceeess.SPComputation and Languagecond-mat.mtrl-sciMultimediacond-mat.mes-halleess.SYInformation Theorymath.ITNetworking and Internet Architecturephysics.app-phphysics.opticsSystems and Control

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
24works
0followers
17topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

24 published item(s)

preprint2026arXiv

Rethinking Secure Semantic Communications in the Age of Generative and Agentic AI: Threats and Opportunities

Semantic communication (SemCom) improves communication efficiency by transmitting task-relevant information instead of raw bits and is expected to be a key technology for 6G networks. Recent advances in generative AI (GenAI) further enhance SemCom by enabling robust semantic encoding and decoding under limited channel conditions. However, these efficiency gains also introduce new security and privacy vulnerabilities. Due to the broadcast nature of wireless channels, eavesdroppers can also use powerful GenAI-based semantic decoders to recover private information from intercepted signals. Moreover, rapid advances in agentic AI enable eavesdroppers to perform long-term and adaptive inference through the integration of memory, external knowledge, and reasoning capabilities. This allows eavesdroppers to further infer user private behavior and intent beyond the transmitted content. Motivated by these emerging challenges, this paper comprehensively rethinks the security and privacy of SemCom systems in the age of generative and agentic AI. We first present a systematic taxonomy of eavesdropping threat models in SemCom systems. Then, we provide insights into how GenAI and agentic AI can enhance eavesdropping threats. Meanwhile, we also highlight potential opportunities for leveraging GenAI and agentic AI to design privacy-preserving SemCom systems.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Step Potential Advantage Estimation: Harnessing Intermediate Confidence and Correctness for Efficient Mathematical Reasoning

Reinforcement Learning with Verifiable Rewards (RLVR) elicits long chain-of-thought reasoning in large language models (LLMs), but outcome-based rewards lead to coarse-grained advantage estimation. While existing approaches improve RLVR via token-level entropy or sequence-level length control, they lack a semantically grounded, step-level measure of reasoning progress. As a result, LLMs fail to distinguish necessary deduction from redundant verification: they may continue checking after reaching a correct solution and, in extreme cases, overturn a correct trajectory into an incorrect final answer. To remedy the lack of process supervision, we introduce a training-free probing mechanism that extracts intermediate confidence and correctness and combines them into a Step Potential signal that explicitly estimates the reasoning state at each step. Building on this signal, we propose Step Potential Advantage Estimation (SPAE), a fine-grained credit assignment method that amplifies potential gains, penalizes potential drops, and applies penalty after potential saturates to encourage timely termination. Experiments across multiple benchmarks show SPAE consistently improves accuracy while substantially reducing response length, outperforming strong RL baselines and recent efficient reasoning and token-level advantage estimation methods. The code is available at https://github.com/cii030/SPAE-RL.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Unlocking Complex Visual Generation via Closed-Loop Verified Reasoning

Despite rapid advancements, current text-to-image (T2I) models predominantly rely on a single-step generation paradigm, which struggles with complex semantics and faces diminishing returns from parameter scaling. While recent multi-step reasoning approaches show promise, they are hindered by ungrounded planning hallucinations lacking verification, monolithic post-hoc reflection, long-context optimization instabilities, and prohibitive inference latency. To overcome these bottlenecks, we propose the Closed-Loop Visual Reasoning (CLVR) framework, a comprehensive system that deeply couples visual-language logical planning with pixel-level diffusion generation. CLVR introduces an automated data engine with step-level visual verification to synthesize reliable reasoning trajectories, and proposes Proxy Prompt Reinforcement Learning (PPRL) to resolve long-context optimization instabilities by distilling interleaved multimodal histories into explicit reward signals for accurate causal attribution. Furthermore, to mitigate the severe latency bottleneck caused by iterative denoising, we propose $Δ$-Space Weight Merge (DSWM), a theoretically grounded method that fuses alignment weights with off-the-shelf distillation priors, reducing the per-step inference cost to just 4 NFEs without requiring expensive re-distillation. Extensive experiments demonstrate that CLVR outperforms existing open-source baselines across multiple benchmarks and approaches the performance of proprietary commercial models, unlocking general test-time scaling capabilities for complex visual generation.

0 citations0 reviewsNo code linkedOpen access
preprint2024arXiv

Continuous Modeling of the Denoising Process for Speech Enhancement Based on Deep Learning

In this paper, we explore a continuous modeling approach for deep-learning-based speech enhancement, focusing on the denoising process. We use a state variable to indicate the denoising process. The starting state is noisy speech and the ending state is clean speech. The noise component in the state variable decreases with the change of the state index until the noise component is 0. During training, a UNet-like neural network learns to estimate every state variable sampled from the continuous denoising process. In testing, we introduce a controlling factor as an embedding, ranging from zero to one, to the neural network, allowing us to control the level of noise reduction. This approach enables controllable speech enhancement and is adaptable to various application scenarios. Experimental results indicate that preserving a small amount of noise in the clean target benefits speech enhancement, as evidenced by improvements in both objective speech measures and automatic speech recognition performance.

0 citations0 reviewsNo code linkedOpen access
preprint2023arXiv

Bidirectional Trained Tree-Structured Decoder for Handwritten Mathematical Expression Recognition

The Handwritten Mathematical Expression Recognition (HMER) task is a critical branch in the field of OCR. Recent studies have demonstrated that incorporating bidirectional context information significantly improves the performance of HMER models. However, existing methods fail to effectively utilize bidirectional context information during the inference stage. Furthermore, current bidirectional training methods are primarily designed for string decoders and cannot adequately generalize to tree decoders, which offer superior generalization capabilities and structural analysis capacity. In order to overcome these limitations, we propose the Mirror-Flipped Symbol Layout Tree (MF-SLT) and Bidirectional Asynchronous Training (BAT) structure. Our method extends the bidirectional training strategy to the tree decoder, allowing for more effective training by leveraging bidirectional information. Additionally, we analyze the impact of the visual and linguistic perception of the HMER model separately and introduce the Shared Language Modeling (SLM) mechanism. Through the SLM, we enhance the model's robustness and generalization when dealing with visual ambiguity, particularly in scenarios with abundant training data. Our approach has been validated through extensive experiments, demonstrating its ability to achieve new state-of-the-art results on the CROHME 2014, 2016, and 2019 datasets, as well as the HME100K dataset. The code used in our experiments will be publicly available.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Lottery Ticket Hypothesis Framework for Low-Complexity Device-Robust Neural Acoustic Scene Classification

We propose a novel neural model compression strategy combining data augmentation, knowledge transfer, pruning, and quantization for device-robust acoustic scene classification (ASC). Specifically, we tackle the ASC task in a low-resource environment leveraging a recently proposed advanced neural network pruning mechanism, namely Lottery Ticket Hypothesis (LTH), to find a sub-network neural model associated with a small amount non-zero model parameters. The effectiveness of LTH for low-complexity acoustic modeling is assessed by investigating various data augmentation and compression schemes, and we report an efficient joint framework for low-complexity multi-device ASC, called \emph{Acoustic Lottery}. Acoustic Lottery could compress an ASC model up to $1/10^{4}$ and attain a superior performance (validation accuracy of 79.4% and Log loss of 0.64) compared to its not compressed seed model. All results reported in this work are based on a joint effort of four groups, namely GT-USTC-UKE-Tencent, aiming to address the "Low-Complexity Acoustic Scene Classification (ASC) with Multiple Devices" in the DCASE 2021 Challenge Task 1a.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Study of Designing Compact Audio-Visual Wake Word Spotting System Based on Iterative Fine-Tuning in Neural Network Pruning

Audio-only-based wake word spotting (WWS) is challenging under noisy conditions due to environmental interference in signal transmission. In this paper, we investigate on designing a compact audio-visual WWS system by utilizing visual information to alleviate the degradation. Specifically, in order to use visual information, we first encode the detected lips to fixed-size vectors with MobileNet and concatenate them with acoustic features followed by the fusion network for WWS. However, the audio-visual model based on neural networks requires a large footprint and a high computational complexity. To meet the application requirements, we introduce a neural network pruning strategy via the lottery ticket hypothesis in an iterative fine-tuning manner (LTH-IF), to the single-modal and multi-modal models, respectively. Tested on our in-house corpus for audio-visual WWS in a home TV scene, the proposed audio-visual system achieves significant performance improvements over the single-modality (audio-only or video-only) system under different noisy conditions. Moreover, LTH-IF pruning can largely reduce the network parameters and computations with no degradation of WWS performance, leading to a potential product solution for the TV wake-up scenario.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A study on joint modeling and data augmentation of multi-modalities for audio-visual scene classification

In this paper, we propose two techniques, namely joint modeling and data augmentation, to improve system performances for audio-visual scene classification (AVSC). We employ pre-trained networks trained only on image data sets to extract video embedding; whereas for audio embedding models, we decide to train them from scratch. We explore different neural network architectures for joint modeling to effectively combine the video and audio modalities. Moreover, data augmentation strategies are investigated to increase audio-visual training set size. For the video modality the effectiveness of several operations in RandAugment is verified. An audio-video joint mixup scheme is proposed to further improve AVSC performances. Evaluated on the development set of TAU Urban Audio Visual Scenes 2021, our final system can achieve the best accuracy of 94.2% among all single AVSC systems submitted to DCASE 2021 Task 1b.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Convergence Theory of Generalized Distributed Subgradient Method with Random Quantization

The distributed subgradient method (DSG) is a widely discussed algorithm to cope with large-scale distributed optimization problems in the arising machine learning applications. Most exisiting works on DSG focus on ideal communication between the cooperative agents such that the shared information between agents is exact and perfect. This assumption, however, could lead to potential privacy concerns and is not feasible when the wireless transmission links are not of good quality. To overcome the challenge, a common approach is to quantize the data locally before transmission, which avoids exposure of raw data and significantly reduces the size of data. Compared with perfect data, quantization poses fundamental challenges on loss of data accuracy, which further impacts the convergence of the algorithms. To settle the problem, we propose a generalized distributed subgradient method with random quantization, which can be intepreted as a two time-scale stochastic approximation method. We provide comprehensive results on the convergence of the algorithm and derive upper bounds on the convergence rates in terms of the quantization bit, stepsizes and the number of network agents. Our results extend the existing results, where only special cases are considered and general conclusions for the convergence rates are missing. Finally, numerical simulations are conducted on linear regression problems to support our theoretical results.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Field-free spin-orbit torque-induced switching of perpendicular magnetization at room temperature in WTe2/ferromagnet heterostructures

Spin-orbit torque (SOT) provides an efficient way to achieve charge-to-spin conversion and can switch perpendicular magnetization, which is essential for designing novel energy-efficient spintronic devices. An out-of-plane SOT could directly switch perpendicular magnetization. Encouragingly, field-free perpendicular magnetization switching of a two-dimensional (2D) material WTe2/ferromagnet (FM) bilayer has been reported recently, but the working temperature (200 K) is below room temperature. Here, we report the field-free perpendicular magnetization switching carried out at room temperature on a WTe2/Pt/Co/Pt multilayer film. Controlled experiments confirm that the field-free switching is caused by the in-plane antidamping SOT generated in the Pt/Co/Pt multilayer and the out-of-plane generated in the a-axis WTe2 thin film. This work offers a potential method for using spintronic devices made of two-dimensional materials at room temperature.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Magnetic phase transition induced ferroelectric polarization in BaFeF4 with room temperature weak ferromagnetism

BaMF4 (M=Fe, Co, Ni and Mn) family are typical multiferroic materials, having antiferromagnetism at around liquid nitrogen temperature. In this work, polycrystalline BaFeF4 has been prepared by solid state reaction. The slight deficiency of Fe leads to the coexistence of valence states of +2 and +3, facilitating the electrons to hop between the neighboring Fe2+ and Fe3+ ions through the middle F- ion, leading to the strong double exchange interaction with weak ferromagnetism above room temperature. A bifurcation at about 170 K between the zero-field-cooled and field-cooled temperature dependent magnetization curves indicates the onset of 2-dimensional antiferromagnetism, which is completed at about 125 K with the sudden drop of magnetization. Despite the fact of type-I multiferroic, its magnetoelectricity can be evidenced by the pyroelectric current, which shows a peak starting at about 170 K and finishing at about 125 K. The saturated ferroelectric polarization change of around 34 μC/m2 is observed, which is switchable by the reversed poling electric field and decreases to about 30 μC/m2 under a magnetic field of 90 kOe. This magnetoelectricity can be qualitatively reproduced by first-principles calculations. Our results represent substantial progress to search for high-temperature multiferroics in ferroelectric fluorides.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Optically Excited Two-Band Amplified Spontaneous Emission from a High-Current-Density Quantum-Dot LED

Laser diodes based on solution-processable materials could benefit numerous technologies including integrated electronics and photonics, telecommunication, and medical diagnostics. An attractive system for implementing these devices is colloidal semiconductor quantum dots (QDs). The primary challenge that hampered progress towards a QD laser diode (QLD) has been fast nonradiative Auger decay of optical-gain-active multicarrier states. Recently, this problem has been resolved by employing continuously graded QDs (cg-QDs) wherein Auger recombination is strongly suppressed. The use of these structures allowed for demonstrations of optical gain with electrical pumping and optically-excited lasing in multilayered LED-like devices. Here we report on achieving the next critical milestone towards a QLD, which is the demonstration of optically excited amplified spontaneous emission from a fully functional high-current density electroluminescent device. This advance has become possible due to excellent optical gain properties of novel 'compact' cg-QDs and a new LED architecture, which allows for concerted optimization of its optical and electrical properties. The results of this work strongly suggest the feasibility of the final step towards a functional QLD, which is the demonstration of lasing with electrical pumping.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Split, embed and merge: An accurate table structure recognizer

Table structure recognition is an essential part for making machines understand tables. Its main task is to recognize the internal structure of a table. However, due to the complexity and diversity in their structure and style, it is very difficult to parse the tabular data into the structured format which machines can understand easily, especially for complex tables. In this paper, we introduce Split, Embed and Merge (SEM), an accurate table structure recognizer. Our model takes table images as input and can correctly recognize the structure of tables, whether they are simple or a complex tables. SEM is mainly composed of three parts, splitter, embedder and merger. In the first stage, we apply the splitter to predict the potential regions of the table row (column) separators, and obtain the fine grid structure of the table. In the second stage, by taking a full consideration of the textual information in the table, we fuse the output features for each table grid from both vision and language modalities. Moreover, we achieve a higher precision in our experiments through adding additional semantic features. Finally, we process the merging of these basic table grids in a self-regression manner. The correspondent merging results is learned through the attention mechanism. In our experiments, SEM achieves an average F1-Measure of 97.11% on the SciTSR dataset which outperforms other methods by a large margin. We also won the first place in the complex table and third place in all tables in ICDAR 2021 Competition on Scientific Literature Parsing, Task-B. Extensive experiments on other publicly available datasets demonstrate that our model achieves state-of-the-art.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

The USTC-Ximalaya system for the ICASSP 2022 multi-channel multi-party meeting transcription (M2MeT) challenge

We propose two improvements to target-speaker voice activity detection (TS-VAD), the core component in our proposed speaker diarization system that was submitted to the 2022 Multi-Channel Multi-Party Meeting Transcription (M2MeT) challenge. These techniques are designed to handle multi-speaker conversations in real-world meeting scenarios with high speaker-overlap ratios and under heavy reverberant and noisy condition. First, for data preparation and augmentation in training TS-VAD models, speech data containing both real meetings and simulated indoor conversations are used. Second, in refining results obtained after TS-VAD based decoding, we perform a series of post-processing steps to improve the VAD results needed to reduce diarization error rates (DERs). Tested on the ALIMEETING corpus, the newly released Mandarin meeting dataset used in M2MeT, we demonstrate that our proposed system can decrease the DER by up to 66.55/60.59% relatively when compared with classical clustering based diarization on the Eval/Test set.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Underwater Differential Game: Finite-Time Target Hunting Task with Communication Delay

This work considers designing an unmanned target hunting system for a swarm of unmanned underwater vehicles (UUVs) to hunt a target with high maneuverability. Differential game theory is used to analyze combat policies of UUVs and the target within finite time. The challenge lies in UUVs must conduct their control policies in consideration of not only the consistency of the hunting team but also escaping behaviors of the target. To obtain stable feedback control policies satisfying Nash equilibrium, we construct the Hamiltonian function with Leibniz's formula. For further taken underwater disturbances and communication delay into consideration, modified deep reinforcement learning (DRL) is provided to investigate the underwater target hunting task in an unknown dynamic environment. Simulations show that underwater disturbances have a large impact on the system considering communication delay. Moreover, consistency tests show that UUVs perform better consistency with a relatively small range of disturbances.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

An Acoustic Segment Model Based Segment Unit Selection Approach to Acoustic Scene Classification with Partial Utterances

In this paper, we propose a sub-utterance unit selection framework to remove acoustic segments in audio recordings that carry little information for acoustic scene classification (ASC). Our approach is built upon a universal set of acoustic segment units covering the overall acoustic scene space. First, those units are modeled with acoustic segment models (ASMs) used to tokenize acoustic scene utterances into sequences of acoustic segment units. Next, paralleling the idea of stop words in information retrieval, stop ASMs are automatically detected. Finally, acoustic segments associated with the stop ASMs are blocked, because of their low indexing power in retrieval of most acoustic scenes. In contrast to building scene models with whole utterances, the ASM-removed sub-utterances, i.e., acoustic utterances without stop acoustic segments, are then used as inputs to the AlexNet-L back-end for final classification. On the DCASE 2018 dataset, scene classification accuracy increases from 68%, with whole utterances, to 72.1%, with segment selection. This represents a competitive accuracy without any data augmentation, and/or ensemble strategy. Moreover, our approach compares favourably to AlexNet-L with attention.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Analyzing Upper Bounds on Mean Absolute Errors for Deep Neural Network Based Vector-to-Vector Regression

In this paper, we show that, in vector-to-vector regression utilizing deep neural networks (DNNs), a generalized loss of mean absolute error (MAE) between the predicted and expected feature vectors is upper bounded by the sum of an approximation error, an estimation error, and an optimization error. Leveraging upon error decomposition techniques in statistical learning theory and non-convex optimization theory, we derive upper bounds for each of the three aforementioned errors and impose necessary constraints on DNN models. Moreover, we assess our theoretical results through a set of image de-noising and speech enhancement experiments. Our proposed upper bounds of MAE for DNN based vector-to-vector regression are corroborated by the experimental results and the upper bounds are valid with and without the "over-parametrization" technique.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Attentive batch normalization for lstm-based acoustic modeling of speech recognition

Batch normalization (BN) is an effective method to accelerate model training and improve the generalization performance of neural networks. In this paper, we propose an improved batch normalization technique called attentive batch normalization (ABN) in Long Short Term Memory (LSTM) based acoustic modeling for automatic speech recognition (ASR). In the proposed method, an auxiliary network is used to dynamically generate the scaling and shifting parameters in batch normalization, and attention mechanisms are introduced to improve their regularized performance. Furthermore, two schemes, frame-level and utterance-level ABN, are investigated. We evaluate our proposed methods on Mandarin and Uyghur ASR tasks, respectively. The experimental results show that the proposed ABN greatly improves the performance of batch normalization in terms of transcription accuracy for both languages.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Challenge of Spatial Cognition for Deep Learning

Given the success of the deep convolutional neural networks (DCNNs) in applications of visual recognition and classification, it would be tantalizing to test if DCNNs can also learn spatial concepts, such as straightness, convexity, left/right, front/back, relative size, aspect ratio, polygons, etc., from varied visual examples of these concepts that are simple and yet vital for spatial reasoning. Much to our dismay, extensive experiments of the type of cognitive psychology demonstrate that the data-driven deep learning (DL) cannot see through superficial variations in visual representations and grasp the spatial concept in abstraction. The root cause of failure turns out to be the learning methodology, not the computational model of the neural network itself. By incorporating task-specific convolutional kernels, we are able to construct DCNNs for spatial cognition tasks that can generalize to input images not drawn from the same distribution of the training set. This work raises a precaution that without manually-incorporated priors or features DCCNs may fail spatial cognitive tasks at rudimentary level.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Correlating Subword Articulation with Lip Shapes for Embedding Aware Audio-Visual Speech Enhancement

In this paper, we propose a visual embedding approach to improving embedding aware speech enhancement (EASE) by synchronizing visual lip frames at the phone and place of articulation levels. We first extract visual embedding from lip frames using a pre-trained phone or articulation place recognizer for visual-only EASE (VEASE). Next, we extract audio-visual embedding from noisy speech and lip videos in an information intersection manner, utilizing a complementarity of audio and visual features for multi-modal EASE (MEASE). Experiments on the TCD-TIMIT corpus corrupted by simulated additive noises show that our proposed subword based VEASE approach is more effective than conventional embedding at the word level. Moreover, visual embedding at the articulation place level, leveraging upon a high correlation between place of articulation and lip shapes, shows an even better performance than that at the phone level. Finally the proposed MEASE framework, incorporating both audio and visual embedding, yields significantly better speech quality and intelligibility than those obtained with the best visual-only and audio-only EASE systems.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Device-Robust Acoustic Scene Classification Based on Two-Stage Categorization and Data Augmentation

In this technical report, we present a joint effort of four groups, namely GT, USTC, Tencent, and UKE, to tackle Task 1 - Acoustic Scene Classification (ASC) in the DCASE 2020 Challenge. Task 1 comprises two different sub-tasks: (i) Task 1a focuses on ASC of audio signals recorded with multiple (real and simulated) devices into ten different fine-grained classes, and (ii) Task 1b concerns with classification of data into three higher-level classes using low-complexity solutions. For Task 1a, we propose a novel two-stage ASC system leveraging upon ad-hoc score combination of two convolutional neural networks (CNNs), classifying the acoustic input according to three classes, and then ten classes, respectively. Four different CNN-based architectures are explored to implement the two-stage classifiers, and several data augmentation techniques are also investigated. For Task 1b, we leverage upon a quantization method to reduce the complexity of two of our top-accuracy three-classes CNN-based architectures. On Task 1a development data set, an ASC accuracy of 76.9\% is attained using our best single classifier and data augmentation. An accuracy of 81.9\% is then attained by a final model fusion of our two-stage ASC classifiers. On Task 1b development data set, we achieve an accuracy of 96.7\% with a model size smaller than 500KB. Code is available: https://github.com/MihawkHu/DCASE2020_task1.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

On Mean Absolute Error for Deep Neural Network Based Vector-to-Vector Regression

In this paper, we exploit the properties of mean absolute error (MAE) as a loss function for the deep neural network (DNN) based vector-to-vector regression. The goal of this work is two-fold: (i) presenting performance bounds of MAE, and (ii) demonstrating new properties of MAE that make it more appropriate than mean squared error (MSE) as a loss function for DNN based vector-to-vector regression. First, we show that a generalized upper-bound for DNN-based vector- to-vector regression can be ensured by leveraging the known Lipschitz continuity property of MAE. Next, we derive a new generalized upper bound in the presence of additive noise. Finally, in contrast to conventional MSE commonly adopted to approximate Gaussian errors for regression, we show that MAE can be interpreted as an error modeled by Laplacian distribution. Speech enhancement experiments are conducted to corroborate our proposed theorems and validate the performance advantages of MAE over MSE for DNN based regression.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Stroke Constrained Attention Network for Online Handwritten Mathematical Expression Recognition

In this paper, we propose a novel stroke constrained attention network (SCAN) which treats stroke as the basic unit for encoder-decoder based online handwritten mathematical expression recognition (HMER). Unlike previous methods which use trace points or image pixels as basic units, SCAN makes full use of stroke-level information for better alignment and representation. The proposed SCAN can be adopted in both single-modal (online or offline) and multi-modal HMER. For single-modal HMER, SCAN first employs a CNN-GRU encoder to extract point-level features from input traces in online mode and employs a CNN encoder to extract pixel-level features from input images in offline mode, then use stroke constrained information to convert them into online and offline stroke-level features. Using stroke-level features can explicitly group points or pixels belonging to the same stroke, therefore reduces the difficulty of symbol segmentation and recognition via the decoder with attention mechanism. For multi-modal HMER, other than fusing multi-modal information in decoder, SCAN can also fuse multi-modal information in encoder by utilizing the stroke based alignments between online and offline modalities. The encoder fusion is a better way for combining multi-modal information as it implements the information interaction one step before the decoder fusion so that the advantages of multiple modalities can be exploited earlier and more adequately when training the encoder-decoder model. Evaluated on a benchmark published by CROHME competition, the proposed SCAN achieves the state-of-the-art performance.

0 citations0 reviewsNo code linkedOpen access
preprint2019arXiv

Attention-based gated scaling adaptative acoustic model for ctc-based speech recognition

In this paper, we propose a novel adaptive technique that uses an attention-based gated scaling (AGS) scheme to improve deep feature learning for connectionist temporal classification (CTC) acoustic modeling. In AGS, the outputs of each hidden layer of the main network are scaled by an auxiliary gate matrix extracted from the lower layer by using attention mechanisms. Furthermore, the auxiliary AGS layer and the main network are jointly trained without requiring second-pass model training or additional speaker information, such as speaker code. On the Mandarin AISHELL-1 datasets, the proposed AGS yields a 7.94% character error rate (CER). To the best of our knowledge, this result is the best recognition accuracy achieved on this dataset by using an end-to-end framework.

0 citations0 reviewsNo code linkedOpen access