December 15, 2015 Pedro Oliveira, Ronan Collobert, Piotr Dollar Paper Learning to Segment Object Candidates In this paper, we propose a new way to generate object proposals, introducing an approach based on a discriminative convolutional network. Our model obtains substantially higher object recall using fewer proposals. We also show that our model is able to generalize to unseen categories it has not seen during training. Areas Artificial Intelligence , Computer Vision , Paper
December 7, 2015 Bolei Zhou, Yuandong Tian, Sainbayar Sukhbaatar, Rob Fergus Paper Simple Bag-of-Words Baseline for Visual Question Answering We describe a very simple bag-of-words baseline for visual question answering. Areas Artificial Intelligence Paper
December 1, 2015 Robert Kraut, Moira Burke Paper Internet Use and Psychological Well-Being: Effects of Activity and Audience Two lines of research fifteen years apart demonstrate that talking with close friends online is associated with improvements in social support, depression, and other measures of well-being. Talking with strangers and reading about acquaintances are not. Areas Data Science , Human Computer Interaction & UX , Paper
November 25, 2015 Tomas Mikolov, Armand Joulin, Marco Baroni Paper A Roadmap Towards Machine Intelligence We describe one possible roadmap how to develop intelligent machines with communication skills that can perform useful tasks for us. Areas Artificial Intelligence Paper
November 23, 2015 Wojciech Zaremba, Tomas Mikolov, Armand Joulin, Rob Fergus Paper Learning Simple Algorithms from Examples We present an approach for learning simple algorithms such as addition or multiplication. Our methods works as a hard attention model on both input and output and it is learn with reinforcement learning. Areas Artificial Intelligence Paper
November 23, 2015 Sainbayar Sukhbaatar, Gabriel Synnaeve, Soumith Chintala, Rob Fergus Paper MazeBase: A Sandbox for Learning from Games Environment for simple 2D maze games, designed as a sandbox for machine learning approaches to reasoning and planning. Areas Artificial Intelligence Paper
November 20, 2015 Marc'Aurelio Ranzato, Sumit Chopra, Michael Auli, Wojciech Zaremba Paper Sequence Level Training with Recurrent Neural Networks This work aims at improving text generation for applications such as summarization, machine translation and image captioning. The key idea is to learn to predict not just the next word but a whole sequence of words, and to train end-to-end at the sequence level. Areas Artificial Intelligence Paper
November 19, 2015 Piotr Bojanowski, Armand Joulin, Tomas Mikolov Paper Alternative Structures for Character-Level RNNs We present two alternative structures to character level recurrent networks. Character level RNNs are known to be very inefficient while achieving lower performance than word level RNNs. Areas Artificial Intelligence Paper
October 4, 2015 Chunqiang (CQ) Tang, Thawan Kooburat, Pradeep Venkat, Akshay Chander, Zhe Wen, Aravind Narayanan, Patrick Dowell, Robert Karl Paper Holistic Configuration Management at Facebook This paper gives a comprehensive description of the use cases, design, implementation, and usage statistics of a suite of tools that manage Facebook’s configuration end-to-end... Areas Systems & Infrastructure Paper
October 4, 2015 Haonan Lu, Kaushik Veeraraghavan, Philippe Ajoux, Jim Hunt, Yee Jiun Song, Wendy Tobagus, Sanjeev Kumar, Wyatt Lloyd Paper Existential Consistency: Measuring and Understanding Consistency at Facebook Replicated storage for large Web services faces a trade-off between stronger forms of consistency and higher performance properties. Stronger consistency prevents anomalies, i.e., unexpected behavior visible to users, and reduces programming complexity. Areas Systems & Infrastructure Paper