arXiv subjects

Tianyi Zhang

Publications and source records attributed to Tianyi Zhang.

At least 307 records · Page 17Linked to original sources

arXiv2023

Knowledge-Based Version Incompatibility Detection for Deep Learning

Version incompatibility issues are rampant when reusing or reproducing deep learning models and applications. Existing techniques are limited to library dependency specifications declared in PyPI. Therefore, these techniques cannot detect version issues due to undocumented version constraints or issues involving hardware drivers or OS. To address this challenge, we propose to leverage the abundant discussions of DL version issues from Stack Overflow to facilitate version incompatibility detection. We reformulate the problem of knowledge extraction as a Question-Answering (QA) problem and use a pre-trained QA model to extract version compatibility knowledge from online discussions. The extracted knowledge is further consolidated into a weighted knowledge graph to detect potential version incompatibilities when reusing a DL project. Our evaluation results show that (1) our approach can accurately extract version knowledge with 84% accuracy, and (2) our approach can accurately identify 65% of known version issues in 10 popular DL projects with a high precision (92%), while two state-of-the-art approaches can only detect 29% and 6% of these issues with 33% and 17% precision respectively.

cs.SE↗

arXiv2023

A Survey of Diffusion Based Image Generation Models: Issues and Their Solutions

Recently, there has been significant progress in the development of large models. Following the success of ChatGPT, numerous language models have been introduced, demonstrating remarkable performance. Similar advancements have also been observed in image generation models, such as Google's Imagen model, OpenAI's DALL-E 2, and stable diffusion models, which have exhibited impressive capabilities in generating images. However, similar to large language models, these models still encounter unresolved challenges. Fortunately, the availability of open-source stable diffusion models and their underlying mathematical principles has enabled the academic community to extensively analyze the performance of current image generation models and make improvements based on this stable diffusion framework. This survey aims to examine the existing issues and the current solutions pertaining to image generation models.

cs.CV↗

arXiv2023

Software Entity Recognition with Noise-Robust Learning

Recognizing software entities such as library names from free-form text is essential to enable many software engineering (SE) technologies, such as traceability link recovery, automated documentation, and API recommendation. While many approaches have been proposed to address this problem, they suffer from small entity vocabularies or noisy training data, hindering their ability to recognize software entities mentioned in sophisticated narratives. To address this challenge, we leverage the Wikipedia taxonomy to develop a comprehensive entity lexicon with 79K unique software entities in 12 fine-grained types, as well as a large labeled dataset of over 1.7M sentences. Then, we propose self-regularization, a noise-robust learning approach, to the training of our software entity recognition (SER) model by accounting for many dropouts. Results show that models trained with self-regularization outperform both their vanilla counterparts and state-of-the-art approaches on our Wikipedia benchmark and two Stack Overflow benchmarks. We release our models, data, and code for future research.

cs.SE↗

arXiv2023

Representing matroids via pasture morphisms

Using the framework of pastures and foundations of matroids developed by Baker-Lorscheid, we give algorithms to: (i) compute the foundation of a matroid, and (ii) compute all morphisms between two pastures. Together, these provide an efficient method of solving many questions of interest in matroid representations, including orientability, non-representability, and computing all representations of a matroid over a finite field.

math.CO↗

arXiv2023

CellMix: A General Instance Relationship based Method for Data Augmentation Towards Pathology Image Classification

In pathology image analysis, obtaining and maintaining high-quality annotated samples is an extremely labor-intensive task. To overcome this challenge, mixing-based methods have emerged as effective alternatives to traditional preprocessing data augmentation techniques. Nonetheless, these methods fail to fully consider the unique features of pathology images, such as local specificity, global distribution, and inner/outer-sample instance relationships. To better comprehend these characteristics and create valuable pseudo samples, we propose the CellMix framework, which employs a novel distribution-oriented in-place shuffle approach. By dividing images into patches based on the granularity of pathology instances and shuffling them within the same batch, the absolute relationships between instances can be effectively preserved when generating new samples. Moreover, we develop a curriculum learning-inspired, loss-driven strategy to handle perturbations and distribution-related noise during training, enabling the model to adaptively fit the augmented data. Our experiments in pathology image classification tasks demonstrate state-of-the-art (SOTA) performance on 7 distinct datasets. This innovative instance relationship-centered method has the potential to inform general data augmentation approaches for pathology image classification. The associated codes are available at https://github.com/sagizty/CellMix.

cs.CV↗

arXiv2023

Decentralized Training of Foundation Models in Heterogeneous Environments

Training foundation models, such as GPT-3 and PaLM, can be extremely expensive, often involving tens of thousands of GPUs running continuously for months. These models are typically trained in specialized clusters featuring fast, homogeneous interconnects and using carefully designed software systems that support both data parallelism and model/pipeline parallelism. Such dedicated clusters can be costly and difficult to obtain. Can we instead leverage the much greater amount of decentralized, heterogeneous, and lower-bandwidth interconnected compute? Previous works examining the heterogeneous, decentralized setting focus on relatively small models that can be trained in a purely data parallel manner. State-of-the-art schemes for model parallel foundation model training, such as Megatron, only consider the homogeneous data center setting. In this paper, we present the first study of training large foundation models with model parallelism in a decentralized regime over a heterogeneous network. Our key technical contribution is a scheduling algorithm that allocates different computational "tasklets" in the training of foundation models to a group of decentralized GPU devices connected by a slow heterogeneous network. We provide a formal cost model and further propose an efficient evolutionary algorithm to find the optimal allocation strategy. We conduct extensive experiments that represent different scenarios for learning over geo-distributed devices simulated using real-world network measurements. In the most extreme case, across 8 different cities spanning 3 continents, our approach is 4.8X faster than prior state-of-the-art training systems (Megatron).

cs.DC↗

arXiv2023

Rapid Image Labeling via Neuro-Symbolic Learning

The success of Computer Vision (CV) relies heavily on manually annotated data. However, it is prohibitively expensive to annotate images in key domains such as healthcare, where data labeling requires significant domain expertise and cannot be easily delegated to crowd workers. To address this challenge, we propose a neuro-symbolic approach called Rapid, which infers image labeling rules from a small amount of labeled data provided by domain experts and automatically labels unannotated data using the rules. Specifically, Rapid combines pre-trained CV models and inductive logic learning to infer the logic-based labeling rules. Rapid achieves a labeling accuracy of 83.33% to 88.33% on four image labeling tasks with only 12 to 39 labeled samples. In particular, Rapid significantly outperforms finetuned CV models in two highly specialized tasks. These results demonstrate the effectiveness of Rapid in learning from small data and its capability to generalize among different tasks. Code and our dataset are publicly available at https://github.com/Neural-Symbolic-Image-Labeling/

cs.CV↗

arXiv2023

The Effect of Masking Strategies on Knowledge Retention by Language Models

Language models retain a significant amount of world knowledge from their pre-training stage. This allows knowledgeable models to be applied to knowledge-intensive tasks prevalent in information retrieval, such as ranking or question answering. Understanding how and which factual information is acquired by our models is necessary to build responsible models. However, limited work has been done to understand the effect of pre-training tasks on the amount of knowledge captured and forgotten by language models during pre-training. Building a better understanding of knowledge acquisition is the goal of this paper. Therefore, we utilize a selection of pre-training tasks to infuse knowledge into our model. In the following steps, we test the model's knowledge retention by measuring its ability to answer factual questions. Our experiments show that masking entities and principled masking of correlated spans based on pointwise mutual information lead to more factual knowledge being retained than masking random tokens. Our findings demonstrate that, like the ability to perform a task, the (factual) knowledge acquired from being trained on that task is forgotten when a model is trained to perform another task (catastrophic forgetting) and how to prevent this phenomenon. To foster reproducibility, the code, as well as the data used in this paper, are openly available.

cs.CL↗

arXiv2023

Automated Summarization of Stack Overflow Posts

Software developers often resort to Stack Overflow (SO) to fill their programming needs. Given the abundance of relevant posts, navigating them and comparing different solutions is tedious and time-consuming. Recent work has proposed to automatically summarize SO posts to concise text to facilitate the navigation of SO posts. However, these techniques rely only on information retrieval methods or heuristics for text summarization, which is insufficient to handle the ambiguity and sophistication of natural language. This paper presents a deep learning based framework called ASSORT for SO post summarization. ASSORT includes two complementary learning methods, ASSORT_S and ASSORT_{IS}, to address the lack of labeled training data for SO post summarization. ASSORT_S is designed to directly train a novel ensemble learning model with BERT embeddings and domainspecific features to account for the unique characteristics of SO posts. By contrast, ASSORT_{IS} is designed to reuse pre-trained models while addressing the domain shift challenge when no training data is present (i.e., zero-shot learning). Both ASSORT_S and ASSORT_{IS} outperform six existing techniques by at least 13% and 7% respectively in terms of the F1 score. Furthermore, a human study shows that participants significantly preferred summaries generated by ASSORT_S and ASSORT_{IS} over the best baseline, while the preference difference between ASSORT_S and ASSORT_{IS} was small.

cs.SE↗

arXiv2023

SAD: Semi-Supervised Anomaly Detection on Dynamic Graphs

Anomaly detection aims to distinguish abnormal instances that deviate significantly from the majority of benign ones. As instances that appear in the real world are naturally connected and can be represented with graphs, graph neural networks become increasingly popular in tackling the anomaly detection problem. Despite the promising results, research on anomaly detection has almost exclusively focused on static graphs while the mining of anomalous patterns from dynamic graphs is rarely studied but has significant application value. In addition, anomaly detection is typically tackled from semi-supervised perspectives due to the lack of sufficient labeled data. However, most proposed methods are limited to merely exploiting labeled data, leaving a large number of unlabeled samples unexplored. In this work, we present semi-supervised anomaly detection (SAD), an end-to-end framework for anomaly detection on dynamic graphs. By a combination of a time-equipped memory bank and a pseudo-label contrastive learning module, SAD is able to fully exploit the potential of large unlabeled samples and uncover underlying anomalies on evolving graph streams. Extensive experiments on four real-world datasets demonstrate that SAD efficiently discovers anomalies from dynamic graphs and outperforms existing advanced methods even when provided with only little labeled data.

cs.LG↗

arXiv2023

Existence of Optical Vortex Solitons in Photorefractive Media

Optical propagation and vortices in nonlinear media have been intensively studied in modern optical physics. In this paper, we establish constraints regarding the propagation constant and provide an existence theory and numerical computations for positive exponentially decaying solutions for a class of ring-profiled solitons in a type of nonlinear media known as a photorefractive nonlinearity. Our methods include constrained minimization and finite element formalism, and we study the vortex profile and its propagation by fixing the energy flux.

math.AP↗

arXiv2023

CDFI: Cross Domain Feature Interaction for Robust Bronchi Lumen Detection

Endobronchial intervention is increasingly used as a minimally invasive means for the treatment of pulmonary diseases. In order to reduce the difficulty of manipulation in complex airway networks, robust lumen detection is essential for intraoperative guidance. However, these methods are sensitive to visual artifacts which are inevitable during the surgery. In this work, a cross domain feature interaction (CDFI) network is proposed to extract the structural features of lumens, as well as to provide artifact cues to characterize the visual features. To effectively extract the structural and artifact features, the Quadruple Feature Constraints (QFC) module is designed to constrain the intrinsic connections of samples with various imaging-quality. Furthermore, we design a Guided Feature Fusion (GFF) module to supervise the model for adaptive feature fusion based on different types of artifacts. Results show that the features extracted by the proposed method can preserve the structural information of lumen in the presence of large visual variations, bringing much-improved lumen detection accuracy.

cs.CV↗

arXiv2023

AutoRepair: Automated Repair for AI-Enabled Cyber-Physical Systems under Safety-Critical Conditions

Cyber-Physical Systems (CPS) have been widely deployed in safety-critical domains such as transportation, power and energy. Recently, there comes an increasing demand in employing deep neural networks (DNNs) in CPS for more intelligent control and decision making in sophisticated industrial safety-critical conditions, giving birth to the class of DNN controllers. However, due to the inherent uncertainty and opaqueness of DNNs, concerns about the safety of DNN-enabled CPS are also surging. In this work, we propose an automated framework named AutoRepair that, given a safety requirement, identifies unsafe control behavior in a DNN controller and repairs them through an optimization-based method. Having an unsafe signal of system execution, AutoRepair iteratively explores the control decision space and searches for the optimal corrections for the DNN controller in order to satisfy the safety requirements. We conduct a comprehensive evaluation of AutoRepair on 6 instances of industry-level DNN-enabled CPS from different safety-critical domains. Evaluation results show that AutoRepair successfully repairs critical safety issues in the DNN controllers, and significantly improves the reliability of CPS.

cs.SE↗

arXiv2023

Giant room-temperature nonlinearities from a monolayer Janus topological semiconductor

Nonlinear optical materials possess wide applications, ranging from terahertz and mid-infrared detection to energy harvesting. Recently, the correlations between nonlinear optical responses and topological properties, such as Berry curvature and the quantum metric tensor, have stimulated great interest. Here, we report giant room-temperature nonlinearities in an emergent non-centrosymmetric two-dimensional topological material, the Janus transition metal dichalcogenides in the 1T' phase, which are synthesized by an advanced atomic-layer substitution method. High harmonic generation, terahertz emission spectroscopy, and second harmonic generation measurements consistently reveal orders-of-the-magnitude enhancement in terahertz-frequency nonlinearities of 1T' MoSSe (e.g., > 50 times higher than 2H MoS$_2$ for 18th order harmonic generation; > 20 times higher than 2H MoS$_2$ for terahertz emission). It is elucidated that such colossal nonlinear optical responses come from topological band mixing and strong inversion symmetry breaking due to the Janus structure. Our work defines general protocols for designing materials with large nonlinearities and preludes the applications of topological materials in optoelectronics down to the monolayer limit. This two-dimensional form of topological materials also constitute a unique platform for examining origin of the anomalous high-harmonic generation, with potential applications as building blocks for scalable attosecond sources.

cond-mat.mes-hall↗

arXiv2023

The $\textit{NuSTAR}$ Extragalactic Surveys: Source Catalogs from the Extended $\textit{Chandra}$ Deep Field-South and the $\textit{Chandra}$ Deep Field-North

We present a routinized and reliable method to obtain source catalogs from the $\textit{Nuclear Spectroscopic Telescope Array}$ ($\textit{NuSTAR}$) extragalactic surveys of the Extended $\textit{Chandra}$ Deep Field-South (E-CDF-S) and $\textit{Chandra}$ Deep Field-North (CDF-N). The $\textit{NuSTAR}$ E-CDF-S survey covers a sky area of $\approx30'\times30'$ to a maximum depth of $\sim$ 230 ks corrected for vignetting in the 3--24 keV band, with a total of 58 sources detected in our E-CDF-S catalog; the $\textit{NuSTAR}$ CDF-N survey covers a sky area of $\approx7'\times10'$ to a maximum depth of $\sim$ 440 ks corrected for vignetting in the 3--24 keV band, with a total of 42 sources detected in our CDF-N catalog that is produced for the first time. We verify the reliability of our two catalogs by crossmatching them with the relevant catalogs from the $\textit{Chandra}$ X-ray observatory, and find that the fluxes of our $\textit{NuSTAR}$ sources are generally consistent with that of their $\textit{Chandra}$ counterparts. Our two catalogs are produced following the exactly same method and made publicly available, thereby providing a uniform platform that facilitates further studies involving these two fields. Our source-detection method provides a systematic approach for source cataloging in other $\textit{NuSTAR}$ extragalactic surveys.

astro-ph.GA↗

arXiv2023

From block-Toeplitz matrices to differential equations on graphs: towards a general theory for scalable masked Transformers

In this paper we provide, to the best of our knowledge, the first comprehensive approach for incorporating various masking mechanisms into Transformers architectures in a scalable way. We show that recent results on linear causal attention (Choromanski et al., 2021) and log-linear RPE-attention (Luo et al., 2021) are special cases of this general mechanism. However by casting the problem as a topological (graph-based) modulation of unmasked attention, we obtain several results unknown before, including efficient d-dimensional RPE-masking and graph-kernel masking. We leverage many mathematical techniques ranging from spectral analysis through dynamic programming and random walks to new algorithms for solving Markov processes on graphs. We provide a corresponding empirical evaluation.

cs.LG↗

arXiv2023

Cost Sharing under Private Valuation and Connection Control

We consider a cost sharing problem on a weighted undirected graph, where all the nodes want to connect to a special node called source, and they need to share the total cost (weights) of the used edges. Each node except for the source has a private valuation of the connection, and it may block others' connections by strategically cutting its adjacent edges to reduce its cost share, which may increase the total cost. We aim to design mechanisms to prevent the nodes from misreporting their valuations and cutting their adjacent edges. We first show that it is impossible for such a mechanism to further satisfy budget balance (cover the total cost) and efficiency (maximize social welfare). Then, we design two feasible cost sharing mechanisms that incentivize each node to offer all its adjacent edges and truthfully report its valuation, and also satisfy either budget balance or efficiency.

cs.GT↗

arXiv2023

DeepLens: Interactive Out-of-distribution Data Detection in NLP Models

Machine Learning (ML) has been widely used in Natural Language Processing (NLP) applications. A fundamental assumption in ML is that training data and real-world data should follow a similar distribution. However, a deployed ML model may suffer from out-of-distribution (OOD) issues due to distribution shifts in the real-world data. Though many algorithms have been proposed to detect OOD data from text corpora, there is still a lack of interactive tool support for ML developers. In this work, we propose DeepLens, an interactive system that helps users detect and explore OOD issues in massive text corpora. Users can efficiently explore different OOD types in DeepLens with the help of a text clustering method. Users can also dig into a specific text by inspecting salient words highlighted through neuron activation analysis. In a within-subjects user study with 24 participants, participants using DeepLens were able to find nearly twice more types of OOD issues accurately with 22% more confidence compared with a variant of DeepLens that has no interaction or visualization support.

cs.HC↗