Tabla de Contenidos: “…8.2 Optimierung: Lernen, um die Kosten zu minimieren -- 8.2.1 Der Gradientenabstieg -- 8.2.2 Die Lernrate -- 8.2.3 Batch-Größe und stochastischer Gradientenabstieg -- Trainingsrunde: -- 8.2.4 Dem lokalen Minimum entkommen -- 8.3 Backpropagation -- 8.4 Die Anzahl der verborgenen Schichten und der Neuronen anpassen -- 8.5 Ein mittleres Netz in Keras -- 8.6 Zusammenfassung -- Schlüsselkonzepte -- 9 Deep Networks verbessern -- 9.1 Die Initialisierung der Gewichte -- 9.1.1 Xavier-Glorot-Verteilungen -- 9.2 Instabile Gradienten -- 9.2.1 Verschwindende Gradienten -- 9.2.2 Explodierende Gradienten -- 9.2.3 Batch-Normalisierung -- 9.3 Modellgeneralisierung (Überanpassung vermeiden) -- 9.3.1 L1- und L2-Regularisierung -- 9.3.2 Dropout -- 9.3.3 Datenaugmentation -- 9.4 Intelligente Optimierer -- 9.4.1 Momentum -- 9.4.2 Nesterov-Momentum -- 9.4.3 AdaGrad -- 9.4.4 AdaDelta und RMSProp -- 9.4.5 Adam -- 9.5 Ein tiefes neuronales Netz in Keras -- 9.6 Regression -- 9.7 TensorBoard -- 9.8 Zusammenfassung -- Schlüsselkonzepte -- Teil III -- Interaktive Anwendungen des Deep Learning -- 10 Maschinelles Sehen -- 10.1 Convolutional Neural Networks -- 10.1.1 Die zweidimensionale Struktur der visuellen Bilddarstellung -- 10.1.2 Berechnungskomplexität -- 10.1.3 Konvolutionsschichten -- 10.1.4 Mehrere Filter -- 10.1.5 Ein Beispiel für Konvolutionsschichten -- 10.2 Hyperparameter von Konvolutionsfiltern -- 10.2.1 Kernel-Größe -- 10.2.2 Schrittlänge -- 10.2.3 Padding -- 10.3 Pooling-Schichten -- 10.4 LeNet-5 in Keras -- 10.5 AlexNet und VGGNet in Keras -- 10.6 Residualnetzwerke -- 10.6.1 Schwindende Gradienten: Das Grauen der tiefen CNN -- 10.6.2 Residualverbindungen -- 10.6.3 ResNet -- 10.7 Anwendungen des maschinellen Sehens -- 10.7.1 Objekterkennung -- R-CNN -- Fast R-CNN -- Faster R-CNN -- YOLO -- 10.7.2 Bildsegmentierung -- Mask R-CNN -- U-Net -- 10.7.3 Transfer-Lernen…”

Tabla de Contenidos: “…3.2.5 Completion of reorganization -- 3.3 Setting up for OLR -- 3.3.1 DBRC setup -- 3.3.2 Output data set creation -- 3.4 Starting online reorganization -- 3.4.1 Initiation -- 3.5 Reorganization process -- 3.6 Normal termination and cleanup -- 3.7 Pausing an online reorganization -- 3.8 OLR commands -- 3.8.1 Start an online reorganization -- 3.8.2 Modify a reorganization in progress -- 3.8.3 Terminate an online reorganization -- 3.8.4 Display reorganizations in progress -- 3.8.5 Database commands -- 3.9 Logging, operations, I/O errors and restart -- 3.9.1 Logging -- 3.9.2 IMS normal termination and restart -- 3.9.3 I/O errors -- 3.9.4 Resuming online reorganization -- 3.9.5 IMS or z/OS failure -- 3.9.6 FDBR with OLR -- 3.10 Utilities -- 3.10.1 Image copy -- 3.10.2 Change accumulation -- 3.10.3 Database recovery -- 3.10.4 Batch backout -- 3.10.5 HD Unload and HD Reload -- 3.11 DBRC changes for OLR -- 3.11.1 RECON records -- 3.11.2 DBRC commands -- 3.11.3 OLR coexistence with IMS Version 7 and IMS Version 8 -- 3.12 Data sharing, FDBR, RSR, and XRF -- 3.12.1 Data sharing support -- 3.12.2 FDBR support -- 3.12.3 XRF support -- 3.12.4 RSR support -- 3.13 OLR performance -- 3.13.1 OSAM sequential buffering -- 3.13.2 Logging -- 3.13.3 Lock contention -- 3.13.4 Buffer pool definitions -- 3.13.5 Buffer contention -- 3.13.6 DASD contention -- 3.13.7 Pacing -- 3.13.8 TCBs used by OLR -- 3.13.9 Executing OLR in a sysplex -- 3.13.10 CPU use with OLR -- Chapter 4. …”

Tabla de Contenidos: “…9.3.7 Update a record using writeback -- 9.3.8 Update a record using a selection key -- 9.3.9 Control Commit and Rollback -- 9.3.10 Call a stored procedure -- 9.3.11 Using database timestamps -- 9.3.12 Execute SQL that you create in your program -- 9.3.13 Work with binary data -- 9.4 Using the Notes Connector -- 9.5 Connection pooling -- 9.6 Performance tips -- 9.6.1 Performance code samples -- 9.7 Tips and tricks -- 9.7.1 Using the Trace metaconnector -- 9.7.2 Map by Name -- 9.7.3 Sorting data -- 9.7.4 Updating key fields -- 9.7.5 Offload database access to server -- 9.8 LEI power programming -- 9.8.1 LEI Programming 1: Transfer data from profile documents -- 9.8.2 LC LSX Example 2: Update Virtual Fields in Notes views -- 9.8.3 LC LSX Example 3: Fix unvirtualized documents -- 9.9 Notes client coding considerations -- 9.10 Other APIs -- 9.10.1 LS:DO -- 9.10.2 JDBC -- 9.10.3 @DbLookup and @DbColumn functions -- 9.11 Additional information about the LC LSX -- Appendix A. …”

Tabla de Contenidos: “…Defining HALDB databases -- 2.1 Overview of HALDB definition -- 2.1.1 Design the logical structure of the database -- 2.1.2 Implement the logical structure with the DBDGEN process -- 2.1.3 Determine the partitioning scheme -- 2.1.4 Create the partitioning scheme -- 2.1.5 Database exit routines -- 2.1.6 System definition -- 2.1.7 Buffer pools -- 2.1.8 Dynamic allocation -- 2.2 DBDGEN differences for HALDB -- Chapter 3. …”

Tabla de Contenidos: “…5.5 PyTorch's autograd: Backpropagating all things -- 5.5.1 Computing the gradient automatically -- 5.5.2 Optimizers a la carte -- 5.5.3 Training, validation, and overfitting -- 5.5.4 Autograd nits and switching it off -- 5.6 Conclusion -- 5.7 Exercise -- 5.8 Summary -- 6 Using a neural network to fit the data -- 6.1 Artificial neurons -- 6.1.1 Composing a multilayer network -- 6.1.2 Understanding the error function -- 6.1.3 All we need is activation -- 6.1.4 More activation functions -- 6.1.5 Choosing the best activation function -- 6.1.6 What learning means for a neural network -- 6.2 The PyTorch nn module -- 6.2.1 Using __call__ rather than forward -- 6.2.2 Returning to the linear model -- 6.3 Finally a neural network -- 6.3.1 Replacing the linear model -- 6.3.2 Inspecting the parameters -- 6.3.3 Comparing to the linear model -- 6.4 Conclusion -- 6.5 Exercises -- 6.6 Summary -- 7 Telling birds from airplanes: Learning from images -- 7.1 A dataset of tiny images -- 7.1.1 Downloading CIFAR-10 -- 7.1.2 The Dataset class -- 7.1.3 Dataset transforms -- 7.1.4 Normalizing data -- 7.2 Distinguishing birds from airplanes -- 7.2.1 Building the dataset -- 7.2.2 A fully connected model -- 7.2.3 Output of a classifier -- 7.2.4 Representing the output as probabilities -- 7.2.5 A loss for classifying -- 7.2.6 Training the classifier -- 7.2.7 The limits of going fully connected -- 7.3 Conclusion -- 7.4 Exercises -- 7.5 Summary -- 8 Using convolutions to generalize -- 8.1 The case for convolutions -- 8.1.1 What convolutions do -- 8.2 Convolutions in action -- 8.2.1 Padding the boundary -- 8.2.2 Detecting features with convolutions -- 8.2.3 Looking further with depth and pooling -- 8.2.4 Putting it all together for our network -- 8.3 Subclassing nn.Module -- 8.3.1 Our network as an nn.Module -- 8.3.2 How PyTorch keeps track of parameters and submodules…”

Tabla de Contenidos: “…7.3.6 Connecting to server: CMBConnection, CMBConnectionPool -- 7.3.7 Schema: CMBSchemaManagement -- 7.3.8 Listing entities: CMBEntity -- 7.3.9 Listing search templates: CMBSearchTemplate -- 7.3.10 Listing attributes: CMBAttribute -- 7.3.11 Listing search criteria: CMBSTCriterion -- 7.3.12 Sorting arrays of beans -- 7.3.13 Conducting a search: CMBQueryServices -- 7.3.14 Processing search results: CMBSearchResults -- 7.3.15 Representing items: CMBItem -- 7.3.16 Displaying item information -- 7.3.17 Managing content: CMBDataManagement -- 7.3.18 Viewing content: CMBDocumentServices -- 7.3.19 Content Manager Version 8 Document Routing system -- 7.4 Conclusion -- Chapter 8. …”

Tabla de Contenidos: “…s SegNet architecture -- 1.7.7 Application of deep learning to object tracking -- 1.7.8 Application of deep learning to texture classification -- 1.7.9 Texture analysis in the world of deep learning -- 1.8 Part G - Summary -- Acknowledgments -- References -- Biographies -- 2 Advanced methods for robust object detection -- 2.1 Introduction -- 2.2 Preliminaries -- 2.3 R-CNN -- 2.3.1 System design -- 2.3.2 Training -- 2.4 SPP-Net -- 2.5 Fast R-CNN -- 2.5.1 Architecture -- 2.5.2 RoI pooling -- 2.5.3 Multitask loss -- 2.5.4 Finetuning strategy -- 2.6 Faster R-CNN -- 2.6.1 Architecture -- 2.6.2 Region proposal networks -- 2.7 Cascade R-CNN -- 2.7.1 Architecture -- 2.7.2 Cascaded bounding box regression -- 2.7.3 Cascaded detection -- 2.8 Multiscale feature representation -- 2.8.1 MS-CNN -- 2.8.1.1 Architecture -- 2.8.2 FPN -- 2.8.2.1 Architecture -- Bottom-up pathway -- Top-down pathway and lateral connections -- 2.9 YOLO -- 2.10 SSD -- 2.10.1 Architecture -- 2.10.2 Training…”

Tabla de Contenidos: “…4 Combating COVID-19 using object detection techniques for next-generation autonomous systems -- 4.1 Introduction -- 4.2 Need for object detection -- 4.3 Object detection techniques -- 4.3.1 R-CNN family -- 4.3.1.1 R-CNN -- 4.3.1.1.1 Network architecture -- 4.3.1.1.2 Advantages -- 4.3.1.1.3 Disadvantages -- 4.3.1.2 Fast R-CNN -- 4.3.1.2.1 Network architecture -- 4.3.1.2.2 The RoI pooling layer -- 4.3.1.2.3 Advantages -- 4.3.1.2.4 Disadvantages -- 4.3.1.3 Faster R-CNN -- 4.3.1.3.1 Network architecture -- 4.3.1.3.2 Advantages -- 4.3.1.3.3 Disadvantages -- 4.3.2 YOLO family -- 4.3.2.1 YOLOv1 -- 4.3.2.1.1 Network architecture -- 4.3.2.1.2 Advantages -- 4.3.2.1.3 Disadvantages -- 4.3.2.2 YOLOv2 -- 4.3.2.2.1 Improvements made over YOLOv1 -- 4.3.2.2.2 Network architecture -- 4.3.2.2.3 Advantages -- 4.3.2.2.4 Disadvantages -- 4.3.2.3 YOLOv3 -- 4.3.2.3.1 Improvements made over YOLOv2 -- 4.3.2.3.2 Network architecture -- 4.3.2.3.3 Advantages -- 4.3.2.3.4 Disadvantages -- 4.4 Applications of objection detection during COVID-19 crisis -- 4.4.1 Module for autonomous systems (pothole detection) -- 4.4.1.1 Architecture -- 4.4.1.2 Results -- 4.4.2 Social distancing detector -- 4.4.2.1 Results -- 4.4.3 COVID-19 detector based on X-rays -- 4.4.3.1 Architecture -- 4.4.3.1.1 Results -- 4.4.4 Face mask detector -- 4.4.4.1 Architecture -- 4.4.4.1.1 Results -- 4.5 Conclusion -- References -- 5 Non-contact measurement system for COVID-19 vital signs to aid mass screening-An alternate approach -- 5.1 Introduction -- 5.2 COVID-19 global scenarios -- 5.2.1 Infections, recovery and mortality rate -- 5.2.2 Economy and environmental impacts -- 5.3 Measurement and testing protocols of COVID-19 -- 5.3.1 Measurement methods -- 5.3.1.1 Pathophysiological tools -- 5.3.1.1.1 Nucleic acid amplification tests -- 5.3.1.1.2 Serological testing -- 5.3.1.2 Physiological assessment tools…”

“…The book provides a comprehensive reference source for the governing class of Great Britain and Ireland from Oliver Cromwell to Winston Churchill, and offers a deep pool of data to support analysis of social, political, economic, and cultural history in the British Isles over the course of more than four centuries…”

“…Private investment in particular is needed to bridge the infrastructure investment gap, given institutional investors' large pools of long-term capital…”

“…These models have been forced to adopt a reductive approach built on the flow of carbon and nutrients between pools that are difficult or even impossible to verify with empirical evidence. …”

“…Each contributed paper was rigorously peer-reviewed by reviewers who were drawn from a large pool of technical committee members as well as other international reviewers in related fields…”

“…Bringing neurodiversity into the larger conversation about workplace diversity, equity, and inclusion is important not only for reasons of fairness: Neurodiverse individuals represent an undervalued talent pool. The authors discuss the challenges neurodivergent employees often face in the workplace and suggest ways that leaders and employees can support their neurodivergent colleagues and help them achieve their potential…”

“…This study examines the kinds of expertise required for successful performance in Navy flag billets, and whether recent pools of officers possess this experience. The authors also examine Navy trends over the past decade to identify the types of expertise likely to become more important for Navy leaders in the future…”

“…To contribute to the existing pool of evidence on the dynamic interplay between illicit trade and armed conflicts, this report looks at illicit trade flows in four separate conflict-affected countries in the MENA region: Iraq, Libya, Syria and Yemen. …”

“…An additional component of this highly available technology is that XSM keeps two identical copies of an independent disk pool at two sites to provide high availability and disaster recovery. …”

“…Soil carbon and nitrogen pools correlated significantly with changes in atmospheric greenhouse gas. …”

“…Each contributed paper was rigorously peer-reviewed by reviewers who were drawn from a large pool of technical committee members as well as other international reviewers in related fields…”

“…Practical experience awaits as you apply object pooling in a real Unity 2D game project, enhancing your game development capabilities. …”