GLORIA

GEOMAR Library Ocean Research Information Access

feed icon rss

Your email was sent successfully. Check your inbox.

An error occurred while sending the email. Please try again.

Proceed reservation?

Export
  • 1
    Keywords: Artificial intelligence. ; Electronic books.
    Type of Medium: Online Resource
    Pages: 1 online resource (562 pages)
    Edition: 1st ed.
    ISBN: 9783031208652
    Series Statement: Lecture Notes in Computer Science Series ; v.13630
    DDC: 006.3
    Language: English
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 2
    Keywords: Artificial intelligence-Design and construction. ; Electronic books.
    Type of Medium: Online Resource
    Pages: 1 online resource (667 pages)
    Edition: 1st ed.
    ISBN: 9783031208683
    Series Statement: Lecture Notes in Computer Science Series ; v.13631
    DDC: 006.3
    Language: English
    Note: Intro -- Preface -- Organization -- Contents - Part III -- Recommender System -- Mixture of Graph Enhanced Expert Networks for Multi-task Recommendation -- 1 Introduction -- 2 The Proposed Method -- 2.1 Deep Interaction Context Exploitation with Multi-channel Graph Neural Network -- 2.2 Graph Enhanced Expert Network -- 2.3 Model Learning -- 2.4 Discussion -- 3 Experiments -- 3.1 Performance Comparison (RQ1) -- 3.2 Effect of the MGNN Module -- 3.3 Study of MoGENet -- 4 Conclusion -- References -- MF-TagRec: Multi-feature Fused Tag Recommendation for GitHub -- 1 Introduction -- 2 Related Work -- 2.1 Tag Recommendation -- 2.2 Tag Recommendation in Open-Source Communities -- 3 Method -- 3.1 Problem Formulation -- 3.2 Overview of MF-TagRec -- 3.3 CNN for Tag Prediction -- 3.4 Network Training Process -- 4 Experiments -- 4.1 Experimental Dataset -- 4.2 Evaluation Metrics -- 4.3 Experimental Settings -- 4.4 Experimental Results -- 5 Conclusions and Future Work -- References -- Co-contrastive Learning for Multi-behavior Recommendation -- 1 Introduction -- 2 Preliminaries -- 3 Methodology -- 3.1 Interactive View Encoder -- 3.2 Fold View Encoder -- 3.3 Divergence Constraint -- 3.4 Co-contrastive Learning -- 3.5 Efficient Joint Learning Without Sampling -- 4 Experiments -- 4.1 Datasets -- 4.2 Compared Models -- 4.3 Experimental Settings -- 4.4 Performance Comparison -- 4.5 Effectiveness Analysis on Data Sparsity Issue -- 4.6 Ablation Study -- 4.7 Parameter Sensitivity Analysis -- 5 Conclusion -- References -- Pattern Matching and Information-Aware Between Reviews and Ratings for Recommendation -- 1 Introduction -- 2 Related Work -- 3 The Proposed MIAN Model -- 3.1 Global Matching Module -- 3.2 Specific Matching Module -- 3.3 Information-Aware Layer -- 3.4 Interaction Aggregation Layer -- 3.5 Joint Learning of Review Matching and Rating Prediction. , 4 Experiments -- 4.1 Experimental Settings -- 4.2 Overall Performance (RQ1) -- 4.3 Ablation Experimental Study (RQ2) -- 4.4 Case Study (RQ3) -- 5 Conclusion -- References -- Cross-View Contrastive Learning for Knowledge-Aware Session-Based Recommendation -- 1 Introduction -- 2 Notations and Problem Statement -- 3 Method -- 3.1 View Generation -- 3.2 Dual-channel Graph View Encoder -- 3.3 Contrastive Learning and Recommendation -- 4 Experiments -- 4.1 Experimental Settings -- 4.2 Over Performance -- 4.3 Model Ablation Study -- 4.4 Handling Different Session Lengths -- 4.5 Hyperparameter Study -- 5 Conclusion -- References -- Reinforcement Learning -- HiSA: Facilitating Efficient Multi-Agent Coordination and Cooperation by Hierarchical Policy with Shared Attention -- 1 Introduction -- 2 Preliminaries and Related Work -- 2.1 Communicative Decentralized Partially Observable Markov Decision Process -- 2.2 Communicative Methods in MAS -- 2.3 Attention Mechanism -- 2.4 Hierarchical Policy with Attention -- 3 Method -- 3.1 Shared Attention Map for Communication -- 3.2 Hierarchical Structure with Shared Attention Mechanism -- 3.3 HiSA for Multi-agent Reinforcement Learning -- 4 Experiments -- 4.1 Experimental Setup -- 4.2 Experiments on the StarCraft Multi-Agent Challenge -- 4.3 Experiments on the Overcooked -- 5 Summary and Outlook -- References -- DDMA: Discrepancy-Driven Multi-agent Reinforcement Learning -- 1 Introduction -- 2 Related Work -- 3 Preliminary -- 3.1 Partially Observable Stochastic Game -- 3.2 Reinforcement Learning -- 4 Method -- 4.1 Initialization of the Multi-agent Policy -- 4.2 Focused Learning of the Multi-agent Policy -- 4.3 Training -- 5 Experiments -- 5.1 Collision Corridor -- 5.2 MPE Scenarios -- 5.3 Ablation Study -- 6 Conclusion -- References -- PRAG: Periodic Regularized Action Gradient for Efficient Continuous Control. , 1 Introduction -- 2 Background -- 3 From TD-Error to Action Gradient Error -- 3.1 TD-Error and TD-Learning -- 3.2 Action Gradient Error and Action Gradient Regularizer -- 3.3 Periodic Regularized Action Gradient Algorithm -- 4 Experiment -- 4.1 Overall Performance -- 4.2 Ablation Study -- 4.3 Parameter Study -- 5 Related Work -- 6 Conclusion -- References -- Identifying Multiple Influential Nodes for Complex Networks Based on Multi-agent Deep Reinforcement Learning -- 1 Introduction -- 2 Problem Formulation -- 3 Multi-agent Identification Framework -- 3.1 General Framework of MAIF -- 3.2 Framework Elements -- 3.3 Independent Actor-Critic Model -- 3.4 Counterfactual Multi-agent (COMA) Policy Gradients and Gate Recurrent Unit (GRU) Network -- 4 Experimental Preliminaries -- 4.1 Dataset -- 4.2 Comparison Method -- 4.3 Susceptible-Infected-Recovered (SIR) Model -- 5 Experimental Results -- 6 Conclusion -- References -- Online Learning in Iterated Prisoner's Dilemma to Mimic Human Behavior -- 1 Introduction -- 2 Related Work -- 3 Background -- 4 Experimental Setup -- 4.1 Iterated Prisoner's Dilemma (IPD) -- 4.2 Behavioral Cloning with Demonstration Rewards -- 4.3 Online Learning Agents -- 5 Results: Algorithms' Tournament -- 5.1 Multi-agent Tournament -- 6 Behavioral Cloning with Human Data -- 7 Clinical Evidences and Implications -- 8 Discussion -- 9 Conclusion -- References -- Optimizing Exploration-Exploitation Trade-off in Continuous Action Spaces via Q-ensemble -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 4 Proposed Method -- 4.1 Ensemble-Based Exploration Strategy -- 4.2 Selective Repeat Update -- 5 Experiments -- 5.1 Parameter Settings -- 5.2 Comparative Evaluation -- 5.3 Ablation Study -- 6 Conclusion -- References -- Hidden Information General Game Playing with Deep Learning and Search -- 1 Introduction -- 2 Background. , 2.1 General Game Playing -- 2.2 Generalised AlphaZero -- 2.3 Recursive Belief-Based Learning -- 3 Method -- 3.1 Propositional Networks for GDL-II -- 3.2 Sampling GDL-II States -- 3.3 CFR Search -- 3.4 Reinforcement Learning -- 4 Experiments -- 4.1 Evaluation Methodology -- 4.2 Results and Discussion -- 5 Conclusion and Future Work -- References -- Sequential Decision Making with ``Sequential Information'' in Deep Reinforcement Learning -- 1 Introduction -- 2 Related Work -- 3 Background -- 3.1 Deep Reinforcement Learning -- 3.2 Depthwise Separable Convolution -- 3.3 3D Temporal Convolution -- 4 Temporal Aggregation Network in DRL -- 5 Experiments -- 5.1 Experimental Setup -- 5.2 Results Analysis -- 6 Conclusion and Future Work -- References -- Two-Stream Communication-Efficient Federated Pruning Network -- 1 Introduction -- 2 Proposed Method -- 2.1 Preliminary -- 2.2 Downstream Compression via DRL Agent -- 2.3 Upstream Compression Based on Proximal Operator -- 3 Experimental Setup and Results -- 3.1 Compared Methods -- 3.2 Datasets and Simulation Settings -- 3.3 Experiment Results -- 3.4 Ablation Study -- 4 Conclusion -- References -- Strong General AI -- Multi-scale Lightweight Neural Network for Real-Time Object Detection -- 1 Introduction -- 2 Related Works -- 3 Methodology -- 3.1 Network Architecture -- 3.2 Fast Down-Sampling Module -- 3.3 Reduced Computational Block -- 3.4 Detection Part -- 4 Experiments -- 4.1 Experiments Setup -- 4.2 Results -- 4.3 Ablation Study -- 5 Conclusion -- References -- Hyperspectral Image Classification Based on Transformer and Generative Adversarial Network -- 1 Introduction -- 2 Related Works -- 2.1 Superpixelwise PCA -- 2.2 Auxiliary Classifier GANs -- 3 Proposed Method -- 3.1 The Framework of the Proposed SPCA-TransGAN -- 3.2 The Network Framework of Generator. , 3.3 The Network Framework of Multi-scale Discriminator -- 4 Experimental Results and Analysis -- 4.1 DataSets -- 4.2 Classification Results on Two Data Sets -- 5 Conclusion -- References -- Deliberation Selector for Knowledge-Grounded Conversation Generation -- 1 Introduction -- 2 Related Work -- 3 Approach -- 3.1 Problem Statement -- 3.2 Model Description -- 4 Experiments -- 4.1 Datasets -- 4.2 Evaluation Metrics -- 4.3 Baselines -- 4.4 Supplementary Details -- 4.5 Experimental Results -- 4.6 Ablation Test -- 4.7 Case Study -- 5 Conclusion -- References -- Training a Lightweight ViT Network for Image Retrieval -- 1 Introduction -- 2 Methodology -- 2.1 Knowledge Distillation with Relaxed Contrastive Loss -- 2.2 Quantized Heterogeneous Knowledge Distillation -- 2.3 Distillation Heterogeneous Quantization for Multi-exit Networks -- 3 Experiments -- 3.1 Experimental Setup -- 3.2 Comprehensive Comparison Results -- 3.3 Analysis of Distillation Quantization of Multi-exit Networks -- 4 Conclusions -- References -- Vision and Perception -- Segmented-Original Image Pairs to Facilitate Feature Extraction in Deep Learning Models -- 1 Introduction -- 2 Method -- 2.1 Datasets -- 2.2 Segmented-Original Image Pair Training Method -- 3 Experiments -- 3.1 Segmentation Algorithm -- 3.2 Classification Tasks -- 3.3 Unsupervised Learning Tasks -- 4 Conclusion -- References -- FusionSeg: Motion Segmentation by Jointly Exploiting Frames and Events -- 1 Introduction -- 2 Related Work -- 2.1 Motion Segmentation -- 2.2 Visual Transformer -- 3 Methodology -- 3.1 Input Representation -- 3.2 Network Architecture -- 3.3 Feature Fusion Method -- 3.4 Multi-Object Association -- 3.5 Feature Matching and Propagation -- 4 Experiment and Results -- 4.1 Implementation Details -- 4.2 Overview of Datasets -- 4.3 Discussion of Results -- 5 Conclusions and Future Work -- References. , Weakly-Supervised Temporal Action Localization with Multi-Head Cross-Modal Attention.
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 3
    Keywords: Artificial intelligence. ; Electronic books.
    Type of Medium: Online Resource
    Pages: 1 online resource (616 pages)
    Edition: 1st ed.
    ISBN: 9783031208621
    Series Statement: Lecture Notes in Computer Science Series ; v.13629
    DDC: 006.3
    Language: English
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 4
    Publication Date: 2024-02-07
    Description: Numerical modeling enables a comprehensive understanding not only of the Earth's system today, but also of the past. To date, a significant amount of time and effort has been devoted to paleoclimate modeling and analysis, which involves the latest and most advanced Paleoclimate Modelling Intercomparison Project phase 4 (PMIP4). The definition of seasonality, which is influenced by slow variations in the Earth's orbital parameters, plays a key role in determining the calculated seasonal cycle of the climate. In contrast to the classical calendar used today, where the lengths of the months and seasons are fixed, the angular calendar calculates the lengths of the months and seasons according to a fixed number of degrees along the Earth's orbit. When comparing simulation results for different time intervals, it is essential to account for the angular calendar to ensure that the data for comparison are from the same position along the Earth's orbit. Most models use the classical calendar, which can lead to strong distortions of the monthly and seasonal values, especially for the climate of the past. Here, by analyzing daily outputs from multiple PMIP4 model simulations, we examine calendar effects on surface air temperature and precipitation under mid-Holocene, Last Interglacial, and pre-industrial climate conditions. We came to the following conclusions. (a) The largest cooling bias occurs in boreal autumn when the classical calendar is applied for the mid-Holocene and Last Interglacial, due to the fact that the vernal equinox is fixed on 21 March. (b) The sign of the temperature anomalies between the Last Interglacial and pre-industrial in boreal autumn can be reversed after the switch from the classical to angular calendar, particularly over the Northern Hemisphere continents. (c) Precipitation over West Africa is overestimated in boreal summer and underestimated in boreal autumn when the classical seasonal cycle is applied. (d) Finally, month-length adjusted values for surface air temperature and precipitation are very similar to the day-length adjusted values, and therefore correcting the calendar based on the monthly model results can largely reduce the artificial bias. In addition, we examine the calendar effects in three transient simulations for 6–0 ka by AWI-ESM, MPI-ESM, and IPSL-CM. We find significant discrepancies between adjusted and unadjusted temperature values over continents for both hemispheres in boreal autumn, while for other seasons the deviations are relatively small. A drying bias can be found in the summer monsoon precipitation in Africa (in the classical calendar), whereby the magnitude of bias becomes smaller over time. Overall, our study underlines the importance of the application of calendar transformation in the analysis of climate simulations. Neglecting the calendar effects could lead to a profound artificial distortion of the calculated seasonal cycle of surface air temperature and precipitation.
    Type: Article , PeerReviewed
    Format: text
    Format: text
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
Close ⊗
This website uses cookies and the analysis tool Matomo. More information can be found here...