PPT-Hierarchical Question-Image Co-Attention

for Visual Question Answering Jiasen Lu Jianwei Yang Dhruv Batra Devi Parikh How to design a model for VQA How to design a model for VQA Recall the image captioning model. Arbelaez. , M. . Maire. , C. . Fowlkes. , J. . Malik. . Contour Detection and Hierarchical image Segmentation. IEEE Trans. on PAMI , 2011. .  . Contour. Detection and Hierarchical Image Segmentation. shan@cs.unc.edu. Clustering Techniques and Applications to Image Segmentation. Roadmap. Unsupervised learning. Clustering categories. Clustering algorithms. K-means. Fuzzy c-means. Kernel-based . Graph-based. Figure 1. Two-level Hierarchical mode Encoder/Decoder Figure 2 shows the block diagram of a three-level hierarchical encoder/decoder. From the original image I, we generate two subsampl . It . does . exactly what the words suggest.. . 1. It is a statement that grabs the reader.. . . 2. It gets their attention.. . . 3. Make the reader want to . . know more.. . It . is a statement(s) that begins a piece of . writing.. 2. It grabs the readers attention.. 3.Makes the reader want to continue reading. .. *. This can also be called a HOOK.*. 6 Types of . Attention . Ling573 . NLP Systems and Applications. April 25, 2013. Deliverable #3. Posted: Code & results due May 10. Focus: Question processing. Classification, reformulation, expansion, . etc. Additional: general improvement motivated by D#2. #N5CR. You’ll be doing this next year!. RUAE. (Close Reading) revision. Concepts 1-7 . LI: Examine in detail the “CONCEPTS OF READING FOR UNDERSTANDING, ANALYSIS AND EVALUATION,” so that we can answer the questions on the . Ling573 . NLP Systems and Applications. April 25, 2013. Deliverable #3. Posted: Code & results due May 10. Focus: Question processing. Classification, reformulation, expansion, . etc. Additional: general improvement motivated by D#2. Lin Ma, . Zhengdong. Lu, and Hang Li. Huawei Noah’s Ark Lab, Hong Kong. http://. www.ee.cuhk.edu.hk. /~lma. /. . Mine the relationships between multiple modalities. Association different modalities. Classification of Transposable Elements . using a Machine . Learning Approach. Introduction. Transposable Elements (TEs) or jumping genes . are DNA . sequences that . have an intrinsic . capability to move within a host genome from one genomic location . Deep Models. Dhruv Batra. Devi Parikh. Aishwarya Agrawal. (EMNLP 2016). Visual Question Answering (VQA). What is Visual Question Answering?. 2. VQA Task. 3. VQA Task. What is the mustache made of?. Watanabe, . Forssman. , Green & . Bohlin. (2012). Aim: Development of cognitive control and its effect on the attention process. Participants: 40 10-month olds and 40 12-month olds. Babies sat on parent’s lap during the experiment. Zheng Li. , Ying Wei, Yu Zhang, Qiang Yang. Hong Kong University of Science and Technology. Cross-Domain Sentiment classification. Training data. Testing data. Books. Restaurant. Sentiment Classifier. Produces a set of . nested clusters . organized as a hierarchical tree. Can be visualized as a . dendrogram. A tree-like diagram that records the sequences of merges or splits. Strengths of Hierarchical Clustering.