[Fri, 13 Nov 2015 ~ Mon, 23 Nov 2015] Deep Learning in arxiv

DEEPREINFORCEMENT LEARNING IN PARAMETERIZED ACTION SPACE

dqn在足球游戏中的研究

代码：Complete source code for our agent isavailable at https://github.com/mhauskn/dqn-hfo and for the RoboCup 2D domainat https://github.com/mhauskn/HFO/.

 

 

DEEPMULTIMODAL SEMANTIC EMBEDDINGS FOR SPEECH AND IMAGES

 

HeterogeneousKnowledge Transfer in Video Emotion Recognition, Attribution and Summarization

 

 

WordEmbedding based Correlation Model for Question/Answer Matching

improvethe matching (word pair) accuracy

 

 

FeatureLearning based Deep Supervised Hashing with Pairwise Labels

1. 网络更深了

2. 前向fc8用了sign

3. 后向用了relex sign

 

 

DeepMetric Learning via Lifted Structured Feature Embedding

 

 

从pair wised-> triplet->(list, lifted structured)

cnn与metric learning更多更深入的结合

 

 

INTEGRATINGDEEP FEATURES FOR MATERIAL RECOGNITION

比较实用的工程技巧

 

 

WIDERFACE: A Face Detection Benchmark

TheWIDER FACE dataset consists of 393, 703 labeled face bounding boxes in 32, 203images

新的人脸数据集

 

Single-viewto Multi-view: Reconstructing Unseen Views with a Convolutional Network

可以在更多需要重构的场景中去尝试

 

TRAININGCNNS WITH LOW-RANK FILTERS FOR EFFICIENT IMAGE CLASSIFICATION

带效果损失的模型大小与前向时间优化

 

 

 

COMPRESSIONOF DEEP CONVOLUTIONAL NEURAL NETWORKS FOR FAST AND LOW POWER MOBILEAPPLICATIONS

模型运行时与大小优化 for 智能设备

 

 

 

 

ImagesDon’t Lie: Transferring Deep Visual Semantic Features to Large-Scale MultimodalLearning to Rank

 

Multimodal Listing Embedding+tuple方式Learning To Rank

目前看到的电商里的第一篇multimodal