Skip to content

Latest commit

 

History

History
34 lines (27 loc) · 5.23 KB

File metadata and controls

34 lines (27 loc) · 5.23 KB

2011

Deep Learning

  • Adaptive Deconvolutional Networks for Mid and High Level Feature Learning. [url]
  • Asymptotic efficiency of deterministic estimators for discrete energy-based models Ratio matching and pseudolikelihood. [url]
  • Building high-level features using large scale unsupervised learning. arxiv
  • Contractive Auto-Encoders-Explicit Invariance During Feature Extraction. [url]
  • Cross-domain activity recognition via transfer learning. [url]
  • Heterogeneous Transfer Learning for Image Classification. [url] ⭐
  • Kinship Verification through Transfer Learning. [url]
  • Learning Deep Energy Models. [url]
  • Learning image representations from the pixel level via hierarchical sparse coding. [url]
  • Natural Language Processing (almost) from Scratch. [url] ⭐
  • On Autoencoders and Score Matching for Energy Based Models. [url]
  • On Optimization Methods for Deep Learning. [url]
  • On the Expressive Power of Deep Architectures. [url]
  • Spike-and-Slab Sparse Coding for Unsupervised Feature Discovery. [url]
  • Unsupervised Learning of Hierarchical Representations with Convolutional Deep Belief Networks. [url]]
  • Unsupervised Models of Images by Spike-and-Slab RBMs. [url]

Attention

  • Learning where to Attend with Deep Architectures for Image Tracking. arxiv

Transfer learning

  • Domain Adaptation for Large-Scale Sentiment Classification - A Deep Learning Approach. [pdf] ⭐
  • Heterogeneous domain adaptation using manifold alignment. [pdf]
  • Heterogeneous Transfer Learning for Image Classification. [pdf] [code]
  • Learning from multiple outlooks. [pdf]
  • Multi-source domain adaptation and its application to early detection of fatigue. [pdf] ⭐
  • What you saw is not what you get: Domain adaptation using asymmetric kernel transforms. [pdf]