LIVIVO - Das Suchportal für Lebenswissenschaften

switch to English language
Zurück zur letzten Suche Suchhistorie
Erweiterte Suche

Ihre letzten Suchen

  1. AU="Gao, Hongyang"
  2. AU="Bedrosian, P. A."
  3. AU="Hotujac, Ljubomir"
  4. AU="Falato, C"
  5. AU="Mayrink, Marcelo Oliveira"
  6. AU="Davidoff, Dahlia F"
  7. AU=Endo A
  8. AU="Chunlei Yu"
  9. AU="Nagao, Asuteka"
  10. AU="Derwael, Céline" AU="Derwael, Céline"
  11. AU="Yao, Yang"
  12. AU=Strutz Frank
  13. AU="Won-Sang Lee"
  14. AU="the ICHseq Investigators" AU="the ICHseq Investigators"
  15. AU="Green, J G"
  16. AU="Xinguang Yang"
  17. AU="Masamed, Rinat"
  18. AU="Flores-Estrada, Diana"
  19. AU="Castellanos, Jason A"
  20. AU="Kiberd, Bryce A"
  21. AU="Kaushal, Deepak"
  22. AU="Rouzaire, Paul"
  23. AU="Mohammed A. S. Abourehab"
  24. AU="Basa, S."
  25. AU="Rohner, Eliane"
  26. AU="Abu-Mahfouz, Adnan M"
  27. AU="Falanti, Andrea"
  28. AU="Yujing Dang"
  29. AU="Clare Duncan"
  30. AU="Calvo Soto, Andrea Patricia"
  31. AU="Joanna I. Olszewska"
  32. AU="Francesco Cavallieri"
  33. AU="Betaieb, Ehssen"
  34. AU="Fan, Xiaoyu"
  35. AU="Riveros-Magaña, Alma Rocío"
  36. AU="Zhang, Wei-Fen"
  37. AU="Ciuca, Catrinel"
  38. AU="Friend, James R"
  39. AU="Colin R. Jackson"
  40. AU="Messina, Claudia"
  41. AU="Faircloth, Chelsey"
  42. AU="Md. Zabirul Islam" AU="Md. Zabirul Islam"
  43. AU="Butcher, Xochitl"
  44. AU="Espay, Alberto J."

Suchergebnis

Treffer 1 - 10 von insgesamt 48

Suchoptionen

  1. Artikel ; Online: A prediction model for assessing hypoglycemia risk in critically ill patients with sepsis.

    Gao, Hongyang / Zhao, Yang

    Heart & lung : the journal of critical care

    2023  Band 62, Seite(n) 43–49

    Abstract: Background: Few studies have reported the risk factors or developed a risk predictive model of hypoglycemia patients with sepsis.: Objective: To develop a predictive model to assess the hypoglycemia risk in critically ill patients with sepsis.: ... ...

    Abstract Background: Few studies have reported the risk factors or developed a risk predictive model of hypoglycemia patients with sepsis.
    Objective: To develop a predictive model to assess the hypoglycemia risk in critically ill patients with sepsis.
    Methods: For this retrospective study, we collected the data from the Medical Information Mart for Intensive Care III and IV (MIMIC-III and MIMIC-IV). All eligible patients from the MIMIC-III were randomly divided into the training set for development of predictive model and testing set for internal validation of the predictive model at a ratio of 8:2. Patients from the MIMIC-IV database were used as the external validation set. The primary endpoint was the occurrence of hypoglycemia. Univariate and multivariate logistic model was used to screen predictors. Adopted receiver operating characteristics (ROC) and calibration curves to estimate the performance of the nomogram.
    Results: The median follow-up time was 5.13 (2.61-9.79) days. Diabetes, dyslipidemia, mean arterial pressure, anion gap, hematocrit, albumin, sequential organ failure assessment, vasopressors, mechanical ventilation and insulin were identified as the predictors for hypoglycemia risk in critically ill patients with sepsis. We constructed a nomogram for predicting hypoglycemia risk in critically ill patients with sepsis based on these predictors. An online individualized predictive tool: https://ghongyang.shinyapps.io/DynNomapp/. The established nomogram had a good predictive ability by ROC and calibration curves in the training set, testing set and external validation cohort.
    Conclusion: A predictive model of hypoglycemia risk was constructed, with a good ability in predicting the risk of hypoglycemia in critically ill patients with sepsis.
    Mesh-Begriff(e) Humans ; Critical Illness ; Retrospective Studies ; Hypoglycemia ; Sepsis/complications ; Critical Care
    Sprache Englisch
    Erscheinungsdatum 2023-06-09
    Erscheinungsland United States
    Dokumenttyp Randomized Controlled Trial ; Journal Article ; Research Support, Non-U.S. Gov't
    ZDB-ID 193129-5
    ISSN 1527-3288 ; 0147-9563
    ISSN (online) 1527-3288
    ISSN 0147-9563
    DOI 10.1016/j.hrtlng.2023.05.010
    Datenquelle MEDical Literature Analysis and Retrieval System OnLINE

    Volltext online

    Zusatzmaterialien

    Kategorien

    Verfügbar in ZB MED Köln/Königswinter

    Zs.A 1035: Hefte anzeigen Standort:
    Je nach Verfügbarkeit (siehe Angabe bei Bestand)
    bis Jg. 1994: Bestellungen von Artikeln über das Online-Bestellformular
    Jg. 1995 - 2021: Lesesall (1.OG)
    ab Jg. 2022: Lesesaal (EG)

    Über subito bestellen

    Dieser Service ist kostenpflichtig (siehe Lieferbedingungen von subito). Bestellungen, die einen Artikel nebst Supplementary Material umfassen, werden grundsätzlich wie mehrfache Bestellungen bearbeitet. Gebühren fallen in diesen Fällen für jede einzelne Bestellung an.

  2. Artikel ; Online: Graph U-Nets.

    Gao, Hongyang / Ji, Shuiwang

    IEEE transactions on pattern analysis and machine intelligence

    2022  Band 44, Heft 9, Seite(n) 4948–4960

    Abstract: We consider the problem of representation learning for graph data. Given images are special cases of graphs with nodes lie on 2D lattices, graph embedding tasks have a natural correspondence with image pixel-wise prediction tasks such as segmentation. ... ...

    Abstract We consider the problem of representation learning for graph data. Given images are special cases of graphs with nodes lie on 2D lattices, graph embedding tasks have a natural correspondence with image pixel-wise prediction tasks such as segmentation. While encoder-decoder architectures like U-Nets have been successfully applied to image pixel-wise prediction tasks, similar methods are lacking for graph data. This is because pooling and up-sampling operations are not natural on graph data. To address these challenges, we propose novel graph pooling and unpooling operations. The gPool layer adaptively selects some nodes to form a smaller graph based on their scalar projection values. We further propose the gUnpool layer as the inverse operation of the gPool layer. Based on our proposed methods, we develop an encoder-decoder model, known as the graph U-Nets. Experimental results on node classification and graph classification tasks demonstrate that our methods achieve consistently better performance than previous models. Along this direction, we extend our methods by integrating attention mechanisms. Based on attention operators, we proposed attention-based pooling and unpooling layers, which can better capture graph topology information. The empirical results on graph classification tasks demonstrate the promising capability of our methods.
    Sprache Englisch
    Erscheinungsdatum 2022-08-04
    Erscheinungsland United States
    Dokumenttyp Journal Article
    ISSN 1939-3539
    ISSN (online) 1939-3539
    DOI 10.1109/TPAMI.2021.3081010
    Datenquelle MEDical Literature Analysis and Retrieval System OnLINE

    Zusatzmaterialien

    Kategorien

    Über subito bestellen

    Dieser Service ist kostenpflichtig (siehe Lieferbedingungen von subito). Bestellungen, die einen Artikel nebst Supplementary Material umfassen, werden grundsätzlich wie mehrfache Bestellungen bearbeitet. Gebühren fallen in diesen Fällen für jede einzelne Bestellung an.

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  3. Artikel: Integrated metabolomic and transcriptomic analyses revealed metabolite variations and regulatory networks in

    Gao, Hongyang / Shi, Min / Zhang, Huiju / Shang, Hongli / Yang, Quan

    Frontiers in plant science

    2024  Band 14, Seite(n) 1325961

    Abstract: To understand the mechanism of the dynamic accumulation of active ingredients ... ...

    Abstract To understand the mechanism of the dynamic accumulation of active ingredients in
    Sprache Englisch
    Erscheinungsdatum 2024-01-10
    Erscheinungsland Switzerland
    Dokumenttyp Journal Article
    ZDB-ID 2613694-6
    ISSN 1664-462X
    ISSN 1664-462X
    DOI 10.3389/fpls.2023.1325961
    Datenquelle MEDical Literature Analysis and Retrieval System OnLINE

    Zusatzmaterialien

    Kategorien

    Über subito bestellen

    Dieser Service ist kostenpflichtig (siehe Lieferbedingungen von subito). Bestellungen, die einen Artikel nebst Supplementary Material umfassen, werden grundsätzlich wie mehrfache Bestellungen bearbeitet. Gebühren fallen in diesen Fällen für jede einzelne Bestellung an.

  4. Buch ; Online: Inferring Data Preconditions from Deep Learning Models for Trustworthy Prediction in Deployment

    Ahmed, Shibbir / Gao, Hongyang / Rajan, Hridesh

    2024  

    Abstract: Deep learning models are trained with certain assumptions about the data during the development stage and then used for prediction in the deployment stage. It is important to reason about the trustworthiness of the model's predictions with unseen data ... ...

    Abstract Deep learning models are trained with certain assumptions about the data during the development stage and then used for prediction in the deployment stage. It is important to reason about the trustworthiness of the model's predictions with unseen data during deployment. Existing methods for specifying and verifying traditional software are insufficient for this task, as they cannot handle the complexity of DNN model architecture and expected outcomes. In this work, we propose a novel technique that uses rules derived from neural network computations to infer data preconditions for a DNN model to determine the trustworthiness of its predictions. Our approach, DeepInfer involves introducing a novel abstraction for a trained DNN model that enables weakest precondition reasoning using Dijkstra's Predicate Transformer Semantics. By deriving rules over the inductive type of neural network abstract representation, we can overcome the matrix dimensionality issues that arise from the backward non-linear computation from the output layer to the input layer. We utilize the weakest precondition computation using rules of each kind of activation function to compute layer-wise precondition from the given postcondition on the final output of a deep neural network. We extensively evaluated DeepInfer on 29 real-world DNN models using four different datasets collected from five different sources and demonstrated the utility, effectiveness, and performance improvement over closely related work. DeepInfer efficiently detects correct and incorrect predictions of high-accuracy models with high recall (0.98) and high F-1 score (0.84) and has significantly improved over prior technique, SelfChecker. The average runtime overhead of DeepInfer is low, 0.22 sec for all unseen datasets. We also compared runtime overhead using the same hardware settings and found that DeepInfer is 3.27 times faster than SelfChecker.

    Comment: Accepted for publication at the 46th International Conference on Software Engineering (ICSE 2024)
    Schlagwörter Computer Science - Software Engineering ; Computer Science - Machine Learning
    Thema/Rubrik (Code) 006
    Erscheinungsdatum 2024-01-25
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  5. Buch ; Online: MotifExplainer

    Yu, Zhaoning / Gao, Hongyang

    a Motif-based Graph Neural Network Explainer

    2022  

    Abstract: We consider the explanation problem of Graph Neural Networks (GNNs). Most existing GNN explanation methods identify the most important edges or nodes but fail to consider substructures, which are more important for graph data. The only method that ... ...

    Abstract We consider the explanation problem of Graph Neural Networks (GNNs). Most existing GNN explanation methods identify the most important edges or nodes but fail to consider substructures, which are more important for graph data. The only method that considers subgraphs tries to search all possible subgraphs and identify the most significant subgraphs. However, the subgraphs identified may not be recurrent or statistically important. In this work, we propose a novel method, known as MotifExplainer, to explain GNNs by identifying important motifs, recurrent and statistically significant patterns in graphs. Our proposed motif-based methods can provide better human-understandable explanations than methods based on nodes, edges, and regular subgraphs. Given an input graph and a pre-trained GNN model, our method first extracts motifs in the graph using well-designed motif extraction rules. Then we generate motif embedding by feeding motifs into the pre-trained GNN. Finally, we employ an attention-based method to identify the most influential motifs as explanations for the final prediction results. The empirical studies on both synthetic and real-world datasets demonstrate the effectiveness of our method.
    Schlagwörter Computer Science - Machine Learning
    Thema/Rubrik (Code) 006
    Erscheinungsdatum 2022-02-01
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  6. Buch ; Online: Molecular Representation Learning via Heterogeneous Motif Graph Neural Networks

    Yu, Zhaoning / Gao, Hongyang

    2022  

    Abstract: We consider feature representation learning problem of molecular graphs. Graph Neural Networks have been widely used in feature representation learning of molecular graphs. However, most existing methods deal with molecular graphs individually while ... ...

    Abstract We consider feature representation learning problem of molecular graphs. Graph Neural Networks have been widely used in feature representation learning of molecular graphs. However, most existing methods deal with molecular graphs individually while neglecting their connections, such as motif-level relationships. We propose a novel molecular graph representation learning method by constructing a heterogeneous motif graph to address this issue. In particular, we build a heterogeneous motif graph that contains motif nodes and molecular nodes. Each motif node corresponds to a motif extracted from molecules. Then, we propose a Heterogeneous Motif Graph Neural Network (HM-GNN) to learn feature representations for each node in the heterogeneous motif graph. Our heterogeneous motif graph also enables effective multi-task learning, especially for small molecular datasets. To address the potential efficiency issue, we propose to use an edge sampler, which can significantly reduce computational resources usage. The experimental results show that our model consistently outperforms previous state-of-the-art models. Under multi-task settings, the promising performances of our methods on combined datasets shed light on a new learning paradigm for small molecular datasets. Finally, we show that our model achieves similar performances with significantly less computational resources by using our edge sampler.

    Comment: International Conference on Machine Learning (2022)
    Schlagwörter Computer Science - Machine Learning
    Erscheinungsdatum 2022-02-01
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  7. Buch ; Online: Gradient Descent Optimizes Infinite-Depth ReLU Implicit Networks with Linear Widths

    Gao, Tianxiang / Gao, Hongyang

    2022  

    Abstract: Implicit deep learning has recently become popular in the machine learning community since these implicit models can achieve competitive performance with state-of-the-art deep networks while using significantly less memory and computational resources. ... ...

    Abstract Implicit deep learning has recently become popular in the machine learning community since these implicit models can achieve competitive performance with state-of-the-art deep networks while using significantly less memory and computational resources. However, our theoretical understanding of when and how first-order methods such as gradient descent (GD) converge on \textit{nonlinear} implicit networks is limited. Although this type of problem has been studied in standard feed-forward networks, the case of implicit models is still intriguing because implicit networks have \textit{infinitely} many layers. The corresponding equilibrium equation probably admits no or multiple solutions during training. This paper studies the convergence of both gradient flow (GF) and gradient descent for nonlinear ReLU activated implicit networks. To deal with the well-posedness problem, we introduce a fixed scalar to scale the weight matrix of the implicit layer and show that there exists a small enough scaling constant, keeping the equilibrium equation well-posed throughout training. As a result, we prove that both GF and GD converge to a global minimum at a linear rate if the width $m$ of the implicit network is \textit{linear} in the sample size $N$, i.e., $m=\Omega(N)$.
    Schlagwörter Computer Science - Machine Learning ; Mathematics - Optimization and Control ; Statistics - Machine Learning
    Thema/Rubrik (Code) 006
    Erscheinungsdatum 2022-05-16
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  8. Artikel: Critical roles of the activation of ethylene pathway genes mediated by DNA demethylation in Arabidopsis hyperhydricity

    Gao, Hongyang / Xia, Xiuying / An, Lijia

    The plant genome. 2022 June, v. 15, no. 2

    2022  

    Abstract: Hyperhydricity (HH) often occurs in plant tissue culture, seriously influencing the commercial micropropagation and genetic improvement. DNA methylation has been studied for its function in plant development and stress responses. However, its potential ... ...

    Abstract Hyperhydricity (HH) often occurs in plant tissue culture, seriously influencing the commercial micropropagation and genetic improvement. DNA methylation has been studied for its function in plant development and stress responses. However, its potential role in HH is unknown. In this study, we report the first comparative DNA methylome analysis of normal and hyperhydric Arabidopsis thaliana (L.) Heynh. seedlings using whole‐genome bisulfite sequencing (BS‐seq). We found that the global methylation level decreased in hyperhydric seedlings, and most of the differentially methylated genes were CHH hypomethylated genes. Moreover, the bisulfite sequencing results showed that hyperhydric seedlings displayed CHH demethylation patterns in the promoter of the ACS1 and ETR1 genes, resulting in upregulated expression of both genes and increased ethylene accumulation. Furthermore, hyperhydric seedling displayed reduced stomatal aperture accompanied by decreased water loss and increased phosphorylation of aquaporins accompanied by increased water uptake. While silver nitrate (AgNO₃) prevented HH by maintained the degree of methylation in the promoter regions of ACS1 and ETR1 and downregulated the transcription of both genes. AgNO₃ also reduced the content of ethylene together with the phosphorylation of aquaporins and water uptake. Taken together, this study suggested that DNA demethylation is a key switch that activates ethylene pathway genes to enable ethylene synthesis and signal transduction, which may subsequently influence aquaporin phosphorylation and stomatal aperture, eventually causing HH; thus, DNA demethylation plays a crucial role in HH. These results provide insights into the epigenetic regulation mechanism of HH and confirm the role of ethylene and AgNO₃ in hyperhydricity control.
    Schlagwörter Arabidopsis thaliana ; DNA ; DNA demethylation ; DNA methylation ; aquaporins ; bisulfites ; epigenetics ; ethylene ; genetic improvement ; hyperhydricity ; micropropagation ; phosphorylation ; plant development ; plant tissues ; seedlings ; signal transduction ; silver nitrate ; stomatal movement ; tissue culture ; water uptake
    Sprache Englisch
    Erscheinungsverlauf 2022-06
    Erscheinungsort John Wiley & Sons, Ltd
    Dokumenttyp Artikel
    Anmerkung JOURNAL ARTICLE
    ZDB-ID 2375444-8
    ISSN 1940-3372 ; 0011-183X
    ISSN (online) 1940-3372
    ISSN 0011-183X
    DOI 10.1002/tpg2.20202
    Datenquelle NAL Katalog (AGRICOLA)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Verfügbar in ZB MED Bonn

    Z 7758: Hefte anzeigen

    Über subito bestellen

    Dieser Service ist kostenpflichtig (siehe Lieferbedingungen von subito). Bestellungen, die einen Artikel nebst Supplementary Material umfassen, werden grundsätzlich wie mehrfache Bestellungen bearbeitet. Gebühren fallen in diesen Fällen für jede einzelne Bestellung an.

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  9. Artikel ; Online: Critical roles of the activation of ethylene pathway genes mediated by DNA demethylation in Arabidopsis hyperhydricity.

    Gao, Hongyang / Xia, Xiuying / An, Lijia

    The plant genome

    2022  Band 15, Heft 2, Seite(n) e20202

    Abstract: Hyperhydricity (HH) often occurs in plant tissue culture, seriously influencing the commercial micropropagation and genetic improvement. DNA methylation has been studied for its function in plant development and stress responses. However, its potential ... ...

    Abstract Hyperhydricity (HH) often occurs in plant tissue culture, seriously influencing the commercial micropropagation and genetic improvement. DNA methylation has been studied for its function in plant development and stress responses. However, its potential role in HH is unknown. In this study, we report the first comparative DNA methylome analysis of normal and hyperhydric Arabidopsis thaliana (L.) Heynh. seedlings using whole-genome bisulfite sequencing (BS-seq). We found that the global methylation level decreased in hyperhydric seedlings, and most of the differentially methylated genes were CHH hypomethylated genes. Moreover, the bisulfite sequencing results showed that hyperhydric seedlings displayed CHH demethylation patterns in the promoter of the ACS1 and ETR1 genes, resulting in upregulated expression of both genes and increased ethylene accumulation. Furthermore, hyperhydric seedling displayed reduced stomatal aperture accompanied by decreased water loss and increased phosphorylation of aquaporins accompanied by increased water uptake. While silver nitrate (AgNO
    Mesh-Begriff(e) Arabidopsis/genetics ; DNA Demethylation ; Epigenesis, Genetic ; Ethylenes/metabolism ; Seedlings ; Water/metabolism
    Chemische Substanzen Ethylenes ; Water (059QF0KO0R)
    Sprache Englisch
    Erscheinungsdatum 2022-03-23
    Erscheinungsland United States
    Dokumenttyp Journal Article ; Research Support, Non-U.S. Gov't
    ZDB-ID 2375444-8
    ISSN 1940-3372 ; 0011-183X
    ISSN (online) 1940-3372
    ISSN 0011-183X
    DOI 10.1002/tpg2.20202
    Datenquelle MEDical Literature Analysis and Retrieval System OnLINE

    Volltext online

    Zusatzmaterialien

    Kategorien

    Verfügbar in ZB MED Bonn

    Z 7758: Hefte anzeigen

    Über subito bestellen

    Dieser Service ist kostenpflichtig (siehe Lieferbedingungen von subito). Bestellungen, die einen Artikel nebst Supplementary Material umfassen, werden grundsätzlich wie mehrfache Bestellungen bearbeitet. Gebühren fallen in diesen Fällen für jede einzelne Bestellung an.

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

  10. Buch ; Online: Wide Neural Networks as Gaussian Processes

    Gao, Tianxiang / Huo, Xiaokai / Liu, Hailiang / Gao, Hongyang

    Lessons from Deep Equilibrium Models

    2023  

    Abstract: Neural networks with wide layers have attracted significant attention due to their equivalence to Gaussian processes, enabling perfect fitting of training data while maintaining generalization performance, known as benign overfitting. However, existing ... ...

    Abstract Neural networks with wide layers have attracted significant attention due to their equivalence to Gaussian processes, enabling perfect fitting of training data while maintaining generalization performance, known as benign overfitting. However, existing results mainly focus on shallow or finite-depth networks, necessitating a comprehensive analysis of wide neural networks with infinite-depth layers, such as neural ordinary differential equations (ODEs) and deep equilibrium models (DEQs). In this paper, we specifically investigate the deep equilibrium model (DEQ), an infinite-depth neural network with shared weight matrices across layers. Our analysis reveals that as the width of DEQ layers approaches infinity, it converges to a Gaussian process, establishing what is known as the Neural Network and Gaussian Process (NNGP) correspondence. Remarkably, this convergence holds even when the limits of depth and width are interchanged, which is not observed in typical infinite-depth Multilayer Perceptron (MLP) networks. Furthermore, we demonstrate that the associated Gaussian vector remains non-degenerate for any pairwise distinct input data, ensuring a strictly positive smallest eigenvalue of the corresponding kernel matrix using the NNGP kernel. These findings serve as fundamental elements for studying the training and generalization of DEQs, laying the groundwork for future research in this area.

    Comment: Accepted by NeurIPS 2023
    Schlagwörter Computer Science - Machine Learning ; Statistics - Machine Learning
    Thema/Rubrik (Code) 006
    Erscheinungsdatum 2023-10-16
    Erscheinungsland us
    Dokumenttyp Buch ; Online
    Datenquelle BASE - Bielefeld Academic Search Engine (Lebenswissenschaftliche Auswahl)

    Volltext online

    Zusatzmaterialien

    Kategorien

    Fernleihe an ZB MED

    Sie können sich den gewünschten Titel als lokale Nutzerin oder lokaler Nutzer von ZB MED direkt an den Standort Köln schicken lassen.

Zum Seitenanfang