Integrate zhouwang/releasenote-update into master

Documentation/current_iteration.md

@@ -90,9 +90,9 @@ Similar to `learning_rate_schedule`, the arguments are interpreted in the same w
 ### New C++ Eval Examples
 The C++ examples [`CNTKLibraryCPPEvalCPUOnlyExamples`](https://github.com/Microsoft/CNTK/tree/release/2.2/Examples/Evaluation/CNTKLibraryCPPEvalCPUOnlyExamples) and [`CNTKLibraryCPPEvalGPUExamples`](https://github.com/Microsoft/CNTK/tree/release/2.2/Examples/Evaluation/CNTKLibraryCPPEvalGPUExamples) illustrate how to use C++ CNTK Library for model evaluation on CPU and GPU. The [UWPImageRecognition](https://github.com/Microsoft/CNTK/tree/release/2.2/Examples/Evaluation/UWPImageRecognition) contains an example using CNTK UWP library for model evaluation.
 ### Add new C# Eval examples
-  * asynchronous evaluation:  [`EvaluationSingleImageAsync()`](https://github.com/Microsoft/CNTK/tree/release/2.2/Examples/Evaluation/CNTKLibraryCSEvalCPUOnlyExamples/CNTKLibraryCSEvalExamples.cs),
-  * evaluating intermediate layers: [`EvaluateIntermediateLayer()`](https://github.com/Microsoft/CNTK/tree/release/2.2/Examples/Evaluation/CNTKLibraryCSEvalCPUOnlyExamples/CNTKLibraryCSEvalExamples.cs),
-  * evaluating outputs from multiple nodes: [`EvaluateCombinedOutputs()`](https://github.com/Microsoft/CNTK/tree/release/2.2/Examples/Evaluation/CNTKLibraryCSEvalCPUOnlyExamples/CNTKLibraryCSEvalExamples.cs).
+  * asynchronous evaluation:  [`EvaluationSingleImageAsync()`](https://github.com/Microsoft/CNTK/tree/release/2.2/Examples/Evaluation/CNTKLibraryCSEvalCPUOnlyExamples/CNTKLibraryCSEvalExamples.cs). Please refer to the section *Run evaluation asynchronously* on the page [Using C#/.NET Managed API] (https://docs.microsoft.com/en-us/cognitive-toolkit/CNTK-Library-Evaluation-on-Windows#using-cnet-managed-api) for details.
+  * evaluating intermediate layers: [`EvaluateIntermediateLayer()`](https://github.com/Microsoft/CNTK/tree/release/2.2/Examples/Evaluation/CNTKLibraryCSEvalCPUOnlyExamples/CNTKLibraryCSEvalExamples.cs)
+  * evaluating outputs from multiple nodes: [`EvaluateCombinedOutputs()`](https://github.com/Microsoft/CNTK/tree/release/2.2/Examples/Evaluation/CNTKLibraryCSEvalCPUOnlyExamples/CNTKLibraryCSEvalExamples.cs)
 
 ## Operations
 ### Noise contrastive estimation node