PPT-RL with subsampling

Theoretical Analysis Motivation Challenges Learning Caching Policies with Subsampling Haonan Wang Hao He Mohammad Alizadeh Hongzi Mao MIT Computer Science and Artificial Intelligence Laboratory. NIbble. 2. Why I’m talking about graphs. Lots of large data . is . graphs. Facebook, Twitter, citation data, and other . social. networks. The web, the blogosphere, the semantic web, Freebase, . W. C. oncepts). TIPL 4700. Presented . & Prepared by Neeraj Gill. 1. Sampling . A. n Analog Signal. 2. Fin is frequency of analog input signal. Fs is sampling frequency of ADC. Nyquist-Shannon sampling theorem: . person 1. person 2. horse 1. horse 2. R-CNN: Regions with CNN features. Input. image. Extract region. proposals (~2k / image). Compute CNN. features. Classify regions. (linear SVM). Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. Quality Assessment of Sequences. Why does quality assessment matter?. DNA -> Data = lots of processes. . => Errors can be introduced. Poor understanding of the data . => Poor Assembly. person. grass. trees. motorbike. road. Evaluation metric. Pixel classification!. Accuracy?. Heavily unbalanced. Common classes are over-emphasized. Intersection over Union. Average across classes and images. Altered time for OH tomorrow: 9:00-10:00 am.. Please complete mid-semester feedback. Semantic Segmentation. The Task. person. grass. trees. motorbike. road. Evaluation metric. Pixel classification!. Accuracy?.