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

Yu Cao

Yu Cao contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Machine LearningComputer VisionArtificial IntelligenceComputation and Languageq-fin.CPq-fin.STq-fin.TReess.IVcond-mat.softDistributed, Parallel, and Cluster ComputingHardware Architecturemath.APmath.NAmath.PRNumerical Analysisphysics.chem-phphysics.comp-ph

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
23works
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

23 published item(s)

preprint2026arXiv

MAD-OPD: Breaking the Ceiling in On-Policy Distillation via Multi-Agent Debate

On-policy distillation (OPD) trains a student on its own trajectories under token-level teacher supervision, but existing methods are capped by a single-teacher capability ceiling: when the teacher errs, the student inherits the error. OPD also remains largely unexplored in agentic tasks, where per-step errors compound across long trajectories and destabilize training. We propose MAD-OPD (Multi-Agent Debate-driven On-Policy Distillation), which breaks this ceiling by recasting the distillation teacher as a deliberative collective of teachers that debate over the student's on-policy state; the debate produces an emergent collective intelligence that supplies token-level supervision, with each teacher's contribution weighted by its post-debate confidence. To extend OPD to agentic tasks, we also introduce On-Policy Agentic Distillation (OPAD), which adds step-level sampling to stabilize training under multi-step error compounding. We additionally derive a task-adaptive divergence principle, selecting JSD (Jensen-Shannon divergence) for agentic stability and reverse KL (Kullback-Leibler) divergence for code generation, and verify it both theoretically and empirically. Across six teacher-student configurations (Qwen3 and Qwen3.5; 1.7B-14B students, 8B-32B teachers) and five agentic and code benchmarks, MAD-OPD ranks first across all six configurations; on the 14B+8B$\to$4B setting it lifts the agentic average by $+2.4\%$ and the code average by $+3.7\%$ over the stronger single-teacher OPD.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Knowledge-Based Decision Support System for In Vitro Fertilization Treatment

In Vitro Fertilization (IVF) is the most widely used Assisted Reproductive Technology (ART). IVF usually involves controlled ovarian stimulation, oocyte retrieval, fertilization in the laboratory with subsequent embryo transfer. The first two steps correspond with follicular phase of females and ovulation in their menstrual cycle. Therefore, we refer to it as the treatment cycle in our paper. The treatment cycle is crucial because the stimulation medications in IVF treatment are applied directly on patients. In order to optimize the stimulation effects and lower the side effects of the stimulation medications, prompt treatment adjustments are in need. In addition, the quality and quantity of the retrieved oocytes have a significant effect on the outcome of the following procedures. To improve the IVF success rate, we propose a knowledge-based decision support system that can provide medical advice on the treatment protocol and medication adjustment for each patient visit during IVF treatment cycle. Our system is efficient in data processing and light-weighted which can be easily embedded into electronic medical record systems. Moreover, an oocyte retrieval oriented evaluation demonstrates that our system performs well in terms of accuracy of advice for the protocols and medications.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Model-Agnostic Data Manipulation Method for Persona-based Dialogue Generation

Towards building intelligent dialogue agents, there has been a growing interest in introducing explicit personas in generation models. However, with limited persona-based dialogue data at hand, it may be difficult to train a dialogue generation model well. We point out that the data challenges of this generation task lie in two aspects: first, it is expensive to scale up current persona-based dialogue datasets; second, each data sample in this task is more complex to learn with than conventional dialogue data. To alleviate the above data issues, we propose a data manipulation method, which is model-agnostic to be packed with any persona-based dialogue generation model to improve its performance. The original training samples will first be distilled and thus expected to be fitted more easily. Next, we show various effective ways that can diversify such easier distilled data. A given base model will then be trained via the constructed data curricula, i.e. first on augmented distilled samples and then on original ones. Experiments illustrate the superiority of our method with two strong base dialogue models (Transformer encoder-decoder and GPT2).

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

COIN: Communication-Aware In-Memory Acceleration for Graph Convolutional Networks

Graph convolutional networks (GCNs) have shown remarkable learning capabilities when processing graph-structured data found inherently in many application areas. GCNs distribute the outputs of neural networks embedded in each vertex over multiple iterations to take advantage of the relations captured by the underlying graphs. Consequently, they incur a significant amount of computation and irregular communication overheads, which call for GCN-specific hardware accelerators. To this end, this paper presents a communication-aware in-memory computing architecture (COIN) for GCN hardware acceleration. Besides accelerating the computation using custom compute elements (CE) and in-memory computing, COIN aims at minimizing the intra- and inter-CE communication in GCN operations to optimize the performance and energy efficiency. Experimental evaluations with widely used datasets show up to 105x improvement in energy consumption compared to state-of-the-art GCN accelerator.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

DeePKS+ABACUS as a Bridge between Expensive Quantum Mechanical Models and Machine Learning Potentials

Recently, the development of machine learning (ML) potentials has made it possible to perform large-scale and long-time molecular simulations with the accuracy of quantum mechanical (QM) models. However, for high-level QM methods, such as density functional theory (DFT) at the meta-GGA level and/or with exact exchange, quantum Monte Carlo, etc., generating a sufficient amount of data for training a ML potential has remained computationally challenging due to their high cost. In this work, we demonstrate that this issue can be largely alleviated with Deep Kohn-Sham (DeePKS), a ML-based DFT model. DeePKS employs a computationally efficient neural network-based functional model to construct a correction term added upon a cheap DFT model. Upon training, DeePKS offers closely-matched energies and forces compared with high-level QM method, but the number of training data required is orders of magnitude less than that required for training a reliable ML potential. As such, DeePKS can serve as a bridge between expensive QM models and ML potentials: one can generate a decent amount of high-accuracy QM data to train a DeePKS model, and then use the DeePKS model to label a much larger amount of configurations to train a ML potential. This scheme for periodic systems is implemented in a DFT package ABACUS, which is open-source and ready for use in various applications.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Dual-CLVSA: a Novel Deep Learning Approach to Predict Financial Markets with Sentiment Measurements

It is a challenging task to predict financial markets. The complexity of this task is mainly due to the interaction between financial markets and market participants, who are not able to keep rational all the time, and often affected by emotions such as fear and ecstasy. Based on the state-of-the-art approach particularly for financial market predictions, a hybrid convolutional LSTM Based variational sequence-to-sequence model with attention (CLVSA), we propose a novel deep learning approach, named dual-CLVSA, to predict financial market movement with both trading data and the corresponding social sentiment measurements, each through a separate sequence-to-sequence channel. We evaluate the performance of our approach with backtesting on historical trading data of SPDR SP 500 Trust ETF over eight years. The experiment results show that dual-CLVSA can effectively fuse the two types of data, and verify that sentiment measurements are not only informative for financial market predictions, but they also contain extra profitable features to boost the performance of our predicting system.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Interpretable Proof Generation via Iterative Backward Reasoning

We present IBR, an Iterative Backward Reasoning model to solve the proof generation tasks on rule-based Question Answering (QA), where models are required to reason over a series of textual rules and facts to find out the related proof path and derive the final answer. We handle the limitations of existed works in two folds: 1) enhance the interpretability of reasoning procedures with detailed tracking, by predicting nodes and edges in the proof path iteratively backward from the question; 2) promote the efficiency and accuracy via reasoning on the elaborate representations of nodes and history paths, without any intermediate texts that may introduce external noise during proof generation. There are three main modules in IBR, QA and proof strategy prediction to obtain the answer and offer guidance for the following procedure; parent node prediction to determine a node in the existing proof that a new child node will link to; child node prediction to find out which new node will be added to the proof. Experiments on both synthetic and paraphrased datasets demonstrate that IBR has better in-domain performance as well as cross-domain transferability than several strong baselines. Our code and models are available at https://github.com/find-knowledge/IBR .

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Phrase-level Textual Adversarial Attack with Label Preservation

Generating high-quality textual adversarial examples is critical for investigating the pitfalls of natural language processing (NLP) models and further promoting their robustness. Existing attacks are usually realized through word-level or sentence-level perturbations, which either limit the perturbation space or sacrifice fluency and textual quality, both affecting the attack effectiveness. In this paper, we propose Phrase-Level Textual Adversarial aTtack (PLAT) that generates adversarial samples through phrase-level perturbations. PLAT first extracts the vulnerable phrases as attack targets by a syntactic parser, and then perturbs them by a pre-trained blank-infilling model. Such flexible perturbation design substantially expands the search space for more effective attacks without introducing too many modifications, and meanwhile maintaining the textual fluency and grammaticality via contextualized generation using surrounding texts. Moreover, we develop a label-preservation filter leveraging the likelihoods of language models fine-tuned on each class, rather than textual similarity, to rule out those perturbations that potentially alter the original class label for humans. Extensive experiments and human evaluation demonstrate that PLAT has a superior attack effectiveness as well as a better label consistency than strong baselines.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Swin-Pose: Swin Transformer Based Human Pose Estimation

Convolutional neural networks (CNNs) have been widely utilized in many computer vision tasks. However, CNNs have a fixed reception field and lack the ability of long-range perception, which is crucial to human pose estimation. Due to its capability to capture long-range dependencies between pixels, transformer architecture has been adopted to computer vision applications recently and is proven to be a highly effective architecture. We are interested in exploring its capability in human pose estimation, and thus propose a novel model based on transformer architecture, enhanced with a feature pyramid fusion structure. More specifically, we use pre-trained Swin Transformer as our backbone and extract features from input images, we leverage a feature pyramid structure to extract feature maps from different stages. By fusing the features together, our model predicts the keypoint heatmap. The experiment results of our study have demonstrated that the proposed transformer-based model can achieve better performance compared to the state-of-the-art CNN-based models.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

CLVSA: A Convolutional LSTM Based Variational Sequence-to-Sequence Model with Attention for Predicting Trends of Financial Markets

Financial markets are a complex dynamical system. The complexity comes from the interaction between a market and its participants, in other words, the integrated outcome of activities of the entire participants determines the markets trend, while the markets trend affects activities of participants. These interwoven interactions make financial markets keep evolving. Inspired by stochastic recurrent models that successfully capture variability observed in natural sequential data such as speech and video, we propose CLVSA, a hybrid model that consists of stochastic recurrent networks, the sequence-to-sequence architecture, the self- and inter-attention mechanism, and convolutional LSTM units to capture variationally underlying features in raw financial trading data. Our model outperforms basic models, such as convolutional neural network, vanilla LSTM network, and sequence-to-sequence model with attention, based on backtesting results of six futures from January 2010 to December 2017. Our experimental results show that, by introducing an approximate posterior, CLVSA takes advantage of an extra regularizer based on the Kullback-Leibler divergence to prevent itself from overfitting traps.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Colorectal Polyp Detection in Real-world Scenario: Design and Experiment Study

Colorectal polyps are abnormal tissues growing on the intima of the colon or rectum with a high risk of developing into colorectal cancer, the third leading cause of cancer death worldwide. Early detection and removal of colon polyps via colonoscopy have proved to be an effective approach to prevent colorectal cancer. Recently, various CNN-based computer-aided systems have been developed to help physicians detect polyps. However, these systems do not perform well in real-world colonoscopy operations due to the significant difference between images in a real colonoscopy and those in the public datasets. Unlike the well-chosen clear images with obvious polyps in the public datasets, images from a colonoscopy are often blurry and contain various artifacts such as fluid, debris, bubbles, reflection, specularity, contrast, saturation, and medical instruments, with a wide variety of polyps of different sizes, shapes, and textures. All these factors pose a significant challenge to effective polyp detection in a colonoscopy. To this end, we collect a private dataset that contains 7,313 images from 224 complete colonoscopy procedures. This dataset represents realistic operation scenarios and thus can be used to better train the models and evaluate a system's performance in practice. We propose an integrated system architecture to address the unique challenges for polyp detection. Extensive experiments results show that our system can effectively detect polyps in a colonoscopy with excellent performance in real time.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Complexity of randomized algorithms for underdamped Langevin dynamics

We establish an information complexity lower bound of randomized algorithms for simulating underdamped Langevin dynamics. More specifically, we prove that the worst $L^2$ strong error is of order $Ω(\sqrt{d}\, N^{-3/2})$, for solving a family of $d$-dimensional underdamped Langevin dynamics, by any randomized algorithm with only $N$ queries to $\nabla U$, the driving Brownian motion and its weighted integration, respectively. The lower bound we establish matches the upper bound for the randomized midpoint method recently proposed by Shen and Lee [NIPS 2019], in terms of both parameters $N$ and $d$.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Financial Markets Prediction with Deep Learning

Financial markets are difficult to predict due to its complex systems dynamics. Although there have been some recent studies that use machine learning techniques for financial markets prediction, they do not offer satisfactory performance on financial returns. We propose a novel one-dimensional convolutional neural networks (CNN) model to predict financial market movement. The customized one-dimensional convolutional layers scan financial trading data through time, while different types of data, such as prices and volume, share parameters (kernels) with each other. Our model automatically extracts features instead of using traditional technical indicators and thus can avoid biases caused by selection of technical indicators and pre-defined coefficients in technical indicators. We evaluate the performance of our prediction model with strictly backtesting on historical trading data of six futures from January 2010 to October 2017. The experiment results show that our CNN model can effectively extract more generalized and informative features than traditional technical indicators, and achieves more robust and profitable financial performance than previous machine learning approaches.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Towards Efficiently Diversifying Dialogue Generation via Embedding Augmentation

Dialogue generation models face the challenge of producing generic and repetitive responses. Unlike previous augmentation methods that mostly focus on token manipulation and ignore the essential variety within a single sample using hard labels, we propose to promote the generation diversity of the neural dialogue models via soft embedding augmentation along with soft labels in this paper. Particularly, we select some key input tokens and fuse their embeddings together with embeddings from their semantic-neighbor tokens. The new embeddings serve as the input of the model to replace the original one. Besides, soft labels are used in loss calculation, resulting in multi-target supervision for a given input. Our experimental results on two datasets illustrate that our proposed method is capable of generating more diverse responses than raw models while remains a similar n-gram accuracy that ensures the quality of generated responses.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

3D Aggregated Faster R-CNN for General Lesion Detection

Lesions are damages and abnormalities in tissues of the human body. Many of them can later turn into fatal diseases such as cancers. Detecting lesions are of great importance for early diagnosis and timely treatment. To this end, Computed Tomography (CT) scans often serve as the screening tool, allowing us to leverage the modern object detection techniques to detect the lesions. However, lesions in CT scans are often small and sparse. The local area of lesions can be very confusing, leading the region based classifier branch of Faster R-CNN easily fail. Therefore, most of the existing state-of-the-art solutions train two types of heterogeneous networks (multi-phase) separately for the candidate generation and the False Positive Reduction (FPR) purposes. In this paper, we enforce an end-to-end 3D Aggregated Faster R-CNN solution by stacking an "aggregated classifier branch" on the backbone of RPN. This classifier branch is equipped with Feature Aggregation and Local Magnification Layers to enhance the classifier branch. We demonstrate our model can achieve the state of the art performance on both LUNA16 and DeepLesion dataset. Especially, we achieve the best single-model FROC performance on LUNA16 with the inference time being 4.2s per processed scan.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

A Deep Reinforcement Learning Approach to Multi-component Job Scheduling in Edge Computing

We are interested in the optimal scheduling of a collection of multi-component application jobs in an edge computing system that consists of geo-distributed edge computing nodes connected through a wide area network. The scheduling and placement of application jobs in an edge system is challenging due to the interdependence of multiple components of each job, and the communication delays between the geographically distributed data sources and edge nodes and their dynamic availability. In this paper we explore the feasibility of applying Deep Reinforcement Learning (DRL) based design to address these challenges. We introduce a DRL actor-critic algorithm that aims to find an optimal scheduling policy to minimize average job slowdown in the edge system. We have demonstrated through simulations that our design outperforms a few existing algorithms, based on both synthetic data and a Google cloud data trace.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

A Progressive Sub-Network Searching Framework for Dynamic Inference

Many techniques have been developed, such as model compression, to make Deep Neural Networks (DNNs) inference more efficiently. Nevertheless, DNNs still lack excellent run-time dynamic inference capability to enable users trade-off accuracy and computation complexity (i.e., latency on target hardware) after model deployment, based on dynamic requirements and environments. Such research direction recently draws great attention, where one realization is to train the target DNN through a multiple-term objective function, which consists of cross-entropy terms from multiple sub-nets. Our investigation in this work show that the performance of dynamic inference highly relies on the quality of sub-net sampling. With objective to construct a dynamic DNN and search multiple high quality sub-nets with minimal searching cost, we propose a progressive sub-net searching framework, which is embedded with several effective techniques, including trainable noise ranking, channel group and fine-tuning threshold setting, sub-nets re-selection. The proposed framework empowers the target DNN with better dynamic inference capability, which outperforms prior works on both CIFAR-10 and ImageNet dataset via comprehensive experiments on different network structures. Taken ResNet18 as an example, our proposed method achieves much better dynamic inference accuracy compared with prior popular Universally-Slimmable-Network by 4.4%-maximally and 2.3%-averagely in ImageNet dataset with the same model size.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Algebraic Bounds on the Rayleigh-Bénard attractor

The Rayleigh-Bénard system with stress-free boundary conditions is shown to have a global attractor in each affine space where velocity has fixed spatial average. The physical problem is shown to be equivalent to one with periodic boundary conditions and certain symmetries. This enables a Gronwall estimate on enstrophy. That estimate is then used to bound the $L^2$ norm of the temperature gradient on the global attractor, which, in turn, is used to find a bounding region for the attractor in the enstrophy, palinstrophy-plane. All final bounds are algebraic in the viscosity and thermal diffusivity, a significant improvement over previously established estimates. The sharpness of the bounds are tested with numerical simulations.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Deep Fashion3D: A Dataset and Benchmark for 3D Garment Reconstruction from Single Images

High-fidelity clothing reconstruction is the key to achieving photorealism in a wide range of applications including human digitization, virtual try-on, etc. Recent advances in learning-based approaches have accomplished unprecedented accuracy in recovering unclothed human shape and pose from single images, thanks to the availability of powerful statistical models, e.g. SMPL, learned from a large number of body scans. In contrast, modeling and recovering clothed human and 3D garments remains notoriously difficult, mostly due to the lack of large-scale clothing models available for the research community. We propose to fill this gap by introducing Deep Fashion3D, the largest collection to date of 3D garment models, with the goal of establishing a novel benchmark and dataset for the evaluation of image-based garment reconstruction systems. Deep Fashion3D contains 2078 models reconstructed from real garments, which covers 10 different categories and 563 garment instances. It provides rich annotations including 3D feature lines, 3D body pose and the corresponded multi-view real images. In addition, each garment is randomly posed to enhance the variety of real clothing deformations. To demonstrate the advantage of Deep Fashion3D, we propose a novel baseline approach for single-view garment reconstruction, which leverages the merits of both mesh and implicit representations. A novel adaptable template is proposed to enable the learning of all types of clothing in a single network. Extensive experiments have been conducted on the proposed dataset to verify its significance and usefulness. We will make Deep Fashion3D publicly available upon publication.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Identification of New Assembly Mode in the Heliconical Nematic Phase via Tender Resonant X-ray Scattering

Helical structures are exciting and are utilized in numerous applications ranging from biotechnology to displays to medicine. Accurate description and understanding of resonance effects in helical structures provides crucial knowledge on molecular packing beyond positional ordering. We exam-ined the manifestation of resonance effects in a nematic phase with heliconical structure, the so called twist bend nematic (NTB) via tender resonant X-ray scattering (TReXS) at the sulfur K-edge. We demonstrate for the first time quantitatively that the energy dependence of the scattering peak in the NTB phase follows the energy dependence of the complex refractive indices measured by X-ray absorption. This allows us to identify a new self-assembly mode for specific sets of liquid crystal dimers in the NTB phase. We anticipate that new avenues in the exploration of complex orientational structures both in static as well as in dynamic modes induced by external stimuli will be pursued.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Pseudo-Labeling for Small Lesion Detection on Diabetic Retinopathy Images

Diabetic retinopathy (DR) is a primary cause of blindness in working-age people worldwide. About 3 to 4 million people with diabetes become blind because of DR every year. Diagnosis of DR through color fundus images is a common approach to mitigate such problem. However, DR diagnosis is a difficult and time consuming task, which requires experienced clinicians to identify the presence and significance of many small features on high resolution images. Convolutional Neural Network (CNN) has proved to be a promising approach for automatic biomedical image analysis recently. In this work, we investigate lesion detection on DR fundus images with CNN-based object detection methods. Lesion detection on fundus images faces two unique challenges. The first one is that our dataset is not fully labeled, i.e., only a subset of all lesion instances are marked. Not only will these unlabeled lesion instances not contribute to the training of the model, but also they will be mistakenly counted as false negatives, leading the model move to the opposite direction. The second challenge is that the lesion instances are usually very small, making them difficult to be found by normal object detectors. To address the first challenge, we introduce an iterative training algorithm for the semi-supervised method of pseudo-labeling, in which a considerable number of unlabeled lesion instances can be discovered to boost the performance of the lesion detector. For the small size targets problem, we extend both the input size and the depth of feature pyramid network (FPN) to produce a large CNN feature map, which can preserve the detail of small lesions and thus enhance the effectiveness of the lesion detector. The experimental results show that our proposed methods significantly outperform the baselines.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Retinopathy of Prematurity Stage Diagnosis Using Object Segmentation and Convolutional Neural Networks

Retinopathy of Prematurity (ROP) is an eye disorder primarily affecting premature infants with lower weights. It causes proliferation of vessels in the retina and could result in vision loss and, eventually, retinal detachment, leading to blindness. While human experts can easily identify severe stages of ROP, the diagnosis of earlier stages, which are the most relevant to determining treatment choice, are much more affected by variability in subjective interpretations of human experts. In recent years, there has been a significant effort to automate the diagnosis using deep learning. This paper builds upon the success of previous models and develops a novel architecture, which combines object segmentation and convolutional neural networks (CNN) to construct an effective classifier of ROP stages 1-3 based on neonatal retinal images. Motivated by the fact that the formation and shape of a demarcation line in the retina is the distinguishing feature between earlier ROP stages, our proposed system first trains an object segmentation model to identify the demarcation line at a pixel level and adds the resulting mask as an additional "color" channel in the original image. Then, the system trains a CNN classifier based on the processed images to leverage information from both the original image and the mask, which helps direct the model's attention to the demarcation line. In a number of careful experiments comparing its performance to previous object segmentation systems and CNN-only systems trained on our dataset, our novel architecture significantly outperforms previous systems in accuracy, demonstrating the effectiveness of our proposed pipeline.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Unsupervised Domain Adaptation on Reading Comprehension

Reading comprehension (RC) has been studied in a variety of datasets with the boosted performance brought by deep neural networks. However, the generalization capability of these models across different domains remains unclear. To alleviate this issue, we are going to investigate unsupervised domain adaptation on RC, wherein a model is trained on labeled source domain and to be applied to the target domain with only unlabeled samples. We first show that even with the powerful BERT contextual representation, the performance is still unsatisfactory when the model trained on one dataset is directly applied to another target dataset. To solve this, we provide a novel conditional adversarial self-training method (CASe). Specifically, our approach leverages a BERT model fine-tuned on the source dataset along with the confidence filtering to generate reliable pseudo-labeled samples in the target domain for self-training. On the other hand, it further reduces domain distribution discrepancy through conditional adversarial learning across domains. Extensive experiments show our approach achieves comparable accuracy to supervised models on multiple large-scale benchmark datasets.

0 citations0 reviewsNo code linkedOpen access