%{search_type} search results

• 1. Why kernels for structured data? 1.1. Supervised machine learning. 1.2. Kernel methods. 1.3. Representing structured data. 1.4. Goals and contributions. 1.5. Outline. 1.6. Bibliographical notes
• 2. Kernel methods in a nutshell. 2.1. Mathematical foundations. 2.2. Recognising patterns with kernels. 2.3. Foundations of kernel methods. 2.4. Kernel machines. 2.5. Summary
• 3. Kernel design. 3.1. General remarks on kernels and examples. 3.2. Kernel functions. 3.3. Introduction to kernels for structured data. 3.4. Prior work. 3.5. Summary
• 4. Basic term kernels. 4.1. Logics for learning. 4.2. Kernels for basic terms. 4.3. Multi-instance learning. 4.4. Related work. 4.5. Applications and experiments
• 5. Graph kernels. 5.1. Motivation and approach. 5.2. Labelled directed graphs. 5.3. Complete graph kernels. 5.4. Walk kernels. 5.5. Cyclic pattern kernels. 5.6. Related work. 5.7. Relational reinforcement learning. 5.8. Molecule classification. 5.9 Summary
• 6. Conclusions.

This book provides a unique treatment of an important area of machine learning and answers the question of how kernel methods can be applied to structured data. Kernel methods are a class of state-of-the-art learning algorithms that exhibit excellent learning results in several application domains. Originally, kernel methods were developed with data in mind that can easily be embedded in a Euclidean vector space. Much real-world data does not have this property but is inherently structured. An example of such data, often consulted in the book, is the (2D) graph structure of molecules formed by their atoms and bonds. The book guides the reader from the basics of kernel methods to advanced algorithms and kernel design for structured data. It is thus useful for readers who seek an entry point into the field as well as experienced researchers.

• Frontmatter; Inhaltsverzeichnis; Vorbemerkungen;
• 1. Einleitung;
• 2. Die Formung des historischen Stoffs in der Johannis; Backmatter.

Thomas Gärtner addresses Johannis, the great epic about the military commander by the late antique poet Corippus. This work is dedicated to the achievements of Johannis Troglita, the general who succeeded in reconquering North Africa under Emperor Justinian I in 548 AD. The present study pursues twofold aims; firstly, it throws light on Corippus? use of historical sources? especially of Procopius. No less significant is the systematic historical location of Corippus? work in the literary tradition of Latin poetry.

• Adversarial Learning
• Image Anomaly Detection with Generative Adversarial Networks
• Image-to-Markup Generation via Paired Adversarial Learning
• Toward an Understanding of Adversarial Examples in Clinical Trials
• ShapeShifter: Robust Physical Adversarial Attack on Faster R-CNN Object Detector
• Anomaly and Outlier Detection
• GridWatch: Sensor Placement and Anomaly Detection in the Electrical Grid
• Incorporating Privileged Information to Unsupervised Anomaly Detection
• L1-Depth Revisited: A Robust Angle-based Outlier Factor in High-dimensional Space
• Beyond Outlier Detection: LookOut for Pictorial Explanation
• Scalable and Interpretable One-class SVMs with Deep Learning and Random Fourier Features
• Group Anomaly Detection using Deep Generative Models
• Applications
• A Discriminative Model for Identifying Readers and Assessing Text Comprehension from Eye Movements
• Face-Cap: Image Captioning using Facial Expression Analysis
• Pedestrian Trajectory Prediction with Structured Memory Hierarchies
• Classification
• Multiple Instance Learning with Bag-level Randomized Trees
• One-class Quantification
• Deep F-Measure Maximization in Multi-Label Classification: A Comparative Study
• Ordinal Label Proportions
• AWX: An Integrated Approach to Hierarchical-Multilabel Classification
• Clustering and Unsupervised Learning
• Clustering in the Presence of Concept Drift
• Time Warp Invariant Dictionary Learning for Time Series Clustering
• How Your Supporters and Opponents Define Your Interestingness
• Deep Learning
• Efficient Decentralized Deep Learning by Dynamic Model Averaging
• Using Supervised Pretraining to Improve Generalization of Neural Networks on Binary Classification Problems
• Towards Efficient Forward Propagation on Resource-Constrained Systems
• Auxiliary Guided Autoregressive Variational Autoencoders
• Cooperative Multi-Agent Policy Gradient
• Parametric t-Distributed Stochastic Exemplar-centered Embedding
• Joint autoencoders: a flexible meta-learning framework
• Privacy Preserving Synthetic Data Release Using Deep Learning
• On Finer Control of Information Flow in LSTMs
• MaxGain: Regularisation of Neural Networks by Constraining Activation Magnitudes
• Ontology alignment based on word embedding and random forest classification
• Domain Adaption in One-Shot Learning
• Ensemble Methods
• Axiomatic Characterization of AdaBoost and the Multiplicative Weight Update Procedure
• Modular Dimensionality Reduction
• Constructive Aggregation and its Application to Forecasting with Dynamic Ensembles
• MetaBags: Bagged Meta-Decision Trees for Regression
• Evaluation
• Visualizing the Feature Importance for Black Box Models
• Efficient estimation of AUC in a sliding window
• Controlling and visualizing the precision-recall tradeoff for external performance indices
• Evaluation Procedures for Forecasting with Spatio-Temporal Data
• A Blended Metric for Multi-label Optimisation and Evaluation.

The three volume proceedings LNAI 11051? 11053 constitutes the refereed proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases, ECML PKDD 2018, held in Dublin, Ireland, in September 2018. The total of 131 regular papers presented in part I and part II was carefully reviewed and selected from 535 submissions; there are 52 papers in the applied data science, nectar and demo track. The contributions were organized in topical sections named as follows: Part I: adversarial learning; anomaly and outlier detection; applications; classification; clustering and unsupervised learning; deep learningensemble methods; and evaluation. Part II: graphs; kernel methods; learning paradigms; matrix and tensor analysis; online and active learning; pattern and sequence mining; probabilistic models and statistical methods; recommender systems; and transfer learning. Part III: ADS data science applications; ADS e-commerce; ADS engineering and design; ADS financial and security; ADS health; ADS sensing and positioning; nectar track; and demo track.

• Graphs.- Temporally Evolving Community Detection and Prediction in Content-Centric Networks.- Local Topological Data Analysis to Uncover the Global Structure of Data Approaching Graph-Structured Topologies.- Similarity Modeling on Heterogeneous Networks via Automatic Path Discovery.- Dynamic hierarchies in temporal directed networks.- Risk-Averse Matchings over Uncertain Graph Databases.- Discovering Urban Travel Demands through Dynamic Zone Correlation in Location-Based Social Networks.- Social-Affiliation Networks: Patterns and the SOAR Model.- ONE-M: Modeling the Co-evolution of Opinions and Network Connections.- Think before You Discard: Accurate Triangle Counting in Graph Streams with Deletions.- Semi-Supervised Blockmodelling with Pairwise Guidance.- Kernel Methods.- Large-scale Nonlinear Variable Selection via Kernel Random Features.- Fast and Provably Effective Multi-view Classification with Landmark-based SVM.- Nystroem-SGD: Fast Learning of Kernel-Classifiers with Conditioned Stochastic Gradient Descent.- Learning Paradigms.- Hyperparameter Learning for Conditional Kernel Mean Embeddings with Rademacher Complexity Bounds.- Deep Learning Architecture Search by Neuro-Cell-based Evolution with Function-Preserving Mutations.- VC-Dimension Based Generalization Bounds for Relational Learning.- Robust Super-Level Set Estimation using Gaussian Processes.- Robust Super-Level Set Estimation using Gaussian Processes.- Scalable Nonlinear AUC Maximization Methods.- Matrix and Tensor Analysis.- Lambert Matrix Factorization.- Identifying and Alleviating Concept Drift in Streaming Tensor Decomposition.- MASAGA: A Linearly-Convergent Stochastic First-Order Method for Optimization on Manifolds.- Block CUR: Decomposing Matrices using Groups of Columns.- Online and Active Learning.- SpectralLeader: Online Spectral Learning for Single Topic Models.- Online Learning of Weighted Relational Rules for Complex Event Recognition.- Toward Interpretable Deep Reinforcement Learning with Linear Model U-Trees.- Online Feature Selection by Adaptive Sub-gradient Methods.- Frame-based Optimal Design.- Hierarchical Active Learning with Proportion Feedback on Regions.- Pattern and Sequence Mining.- An Efficient Algorithm for Computing Entropic Measures of Feature Subsets.- Anytime Subgroup Discovery in Numerical Domains with Guarantees.- Discovering Spatio-Temporal Latent Influence in Geographical Attention Dynamics.- Mining Periodic Patterns with a MDL Criterion.- Revisiting Conditional Functional Dependency Discovery: Splitting the "C" from the "FD".- Sqn2Vec: Learning Sequence Representation via Sequential Patterns with a Gap Constraint.- Mining Tree Patterns with Partially Injective Homomorphisms.- Probabilistic Models and Statistical Methods.- Variational Bayes for Mixture Models with Censored Data.- Exploration Enhanced Expected Improvement for Bayesian Optimization.- A Left-to-right Algorithm for Likelihood Estimation in Gamma-Poisson Factor Analysis.- Causal Inference on Multivariate and Mixed-Type Data.- Recommender Systems.- POLAR: Attention-based CNN for One-shot Personalized Article Recommendation.- Learning Multi-granularity Dynamic Network Representations for Social Recommendation.- GeoDCF: Deep Collaborative Filtering with Multifaceted Contextual Information in Location-based Social Networks.- Personalized Thread Recommendation for MOOC Discussion Forums.- Inferring Continuous Latent Preference on Transition Intervals for Next Point-of-Interest Recommendation.- Transfer Learning.- Feature Selection for Unsupervised Domain Adaptation using Optimal Transport.- Towards more Reliable Transfer Learning.- Differentially Private Hypothesis Transfer Learning.- Information-theoretic Transfer Learning framework for Bayesian Optimisation.- A Unified Framework for Domain Adaptation using Metric Learning on Manifolds.
• (source: Nielsen Book Data)

The three volume proceedings LNAI 11051 - 11053 constitutes the refereed proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases, ECML PKDD 2018, held in Dublin, Ireland, in September 2018. The total of 131 regular papers presented in part I and part II was carefully reviewed and selected from 535 submissions; there are 52 papers in the applied data science, nectar and demo track. The contributions were organized in topical sections named as follows: Part I: adversarial learning; anomaly and outlier detection; applications; classification; clustering and unsupervised learning; deep learningensemble methods; and evaluation. Part II: graphs; kernel methods; learning paradigms; matrix and tensor analysis; online and active learning; pattern and sequence mining; probabilistic models and statistical methods; recommender systems; and transfer learning. Part III: ADS data science applications; ADS e-commerce; ADS engineering and design; ADS financial and security; ADS health; ADS sensing and positioning; nectar track; and demo track.
(source: Nielsen Book Data)