作者：chen_h
微信号 & QQ：862251340
微信公众号：coderpai
1.【论文 & 代码】Inferring and Executing Programs for Visual Reasoning
简介：
Existing methods for visual reasoning attempt to directly map inputs to outputs using black-box architectures without explicitly modeling the underlying reasoning processes. As a result, these black-box models often learn to exploit biases in the data rather than learning to perform visual reasoning. Inspired by module networks, this paper proposes a model for visual reasoning that consists of a program generator that constructs an explicit representation of the reasoning process to be performed, and an execution engine that executes the resulting program to produce an answer. Both the program generator and the execution engine are implemented by neural networks, and are trained using a combination of backpropagation and REINFORCE. Using the CLEVR benchmark for visual reasoning, we show that our model significantly outperforms strong baselines and generalizes better in a variety of settings.
原文链接：https://github.com/facebookresearch/clevr-iep
2.【代码】Deep Feature Flow for Video Recognition
简介：
Deep Feature Flow is initially described in a CVPR 2017 paper. It provides a simple, fast, accurate, and end-to-end framework for video recognition (e.g., object detection and semantic segmentation in videos). It is worth noting that:
* Deep Feature Flow significantly speeds up video recognition by applying the heavy-weight image recognition network (e.g., ResNet-101) on sparse key frames, and propagating the recognition outputs (feature maps) to the other frames by the light-weight flow network (e.g., FlowNet).
* The entire system is end-to-end trained for the task of video recognition, which is vital for improving the recognition accuracy. Directly adopting state-of-the-art flow estimation methods without end-to-end training would deliver noticable worse results.
* Deep Feature Flow can easily make use of sparsely annotated video recognition datasets, where only a small portion of the frames are annotated with ground-truth labels.
原文链接：https://github.com/msracver/Deep-Feature-Flow
3.【博客】Learn Python for Data Science from Scratch
简介：
Python is a multipurpose programming language and widely used for Data Science, which is termed as the sexiest job of this century. Data Scientist mine thru the large dataset to gain insight and make meaningful data driven decisions. Python is used as general purposed programming language and used for Web Development, Networking, Scientific computing etc. We will be discussing further about the series of awesome libraries in python such as numpy, scipy & pandas for data manipulation & wrangling and matplotlib, seaborn & bokeh for data visualization.
原文链接：http://www.datasciencecentral.com/profiles/blogs/learn-python-for-data-science-from-scratch
4.【博客】Keras Tensorflow tutorial: Practical guide from getting started to developing complex deep neural network
简介：
In this quick tutorial, we shall learn following things:
1. Why Keras? Why is it considered to be the future of deep learning?
2. Installing Keras on Ubuntu: Step by step installation on Ubuntu
3. Keras Tensorflow tutorial: Fundamentals of Keras
4. Understanding Keras Sequential Model
4.1) Solve a linear regression problem with example
5. Saving and restoring pre-trained models using Keras
6. Keras functional API
6.1) Develop VGG convolutional neural network using functional API
6.2) Build and run SqueezeNet convolutional neural network using functional API
原文链接：http://cv-tricks.com/tensorflow-tutorial/keras/
5.【论文】The Next Generation Neural Networks: Deep Learning and
Spiking Neural Networks
简介：
Deep Learning and Spike Neural Networks are hot topics in artificial intelligence and human brain. By explaining the basic underlying blocks beneath them, the architectures and applications of both concepts are discovered.
原文链接：https://www.nst.ei.tum.de/fileadmin/w00bqs/www/publications/as/2014SS-HS-DeepLearningAndSNN.pdf
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
