Artificial Intelligence is advancing in an incredible fast pace and staying up to date with the state-of-the-art research is, sometimes, overwhelming. This repository is my "reading list", a collection of interesting papers, courses, blogs, videos and other resources related to Machine Learning and Cognitive Systems in general.
-
Building Machines That Learn and Think Like People (2016) - One one my favorite papers. Discusses the elements of human cognitition that are still missing in current Deep Learning and Reinforcement Learning approaches.
-
Human-level concept learning through probabilistic program induction - A different approach to learning concepts from few data, like humans do.
-
DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition - "We evaluate whether features extracted from the activation of a deep convolutional network trained in a fully supervised fashion on a large, fixed set of object recognition tasks can be repurposed to novel generic tasks".
-
Deep Residual Learning for Image Recognition (2015) - "Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously".
-
A Neural Algorithm of Artistic Style (2015) - "Here we introduce an artificial system based on a Deep Neural Network that creates artistic images of high perceptual quality. The system uses neural representations to separate and recombine content and style of arbitrary images, providing a neural algorithm for the creation of artistic images".
-
ImageNet Classification with Deep Convolutional Neural Networks (2012) - "We trained a large, deep convolutional neural network to classify the 1.3 million high-resolution images in the LSVRC-2010 ImageNet training set into the 1000 different classes".
-
See, Hear, and Read: Deep Aligned Representations (2017) - "We capitalize on large amounts of readily-available, synchronous data to learn a deep discriminative representations shared across three major natural modalities: vision, sound and language. By leveraging over a year of sound from video and millions of sentences paired with images, we jointly train a deep convolutional network for aligned representation learning."
-
Look, Listen and Learn (2017) - "We consider the question: what can be learnt by looking at and listening to a large amount of unlabelled videos? There is a valuable, but so far untapped, source of information contained in the video itself -- the correspondence between the visual and the audio streams, and we introduce a novel "Audio-Visual Correspondence" learning task that makes use of this. Training visual and audio networks from scratch, without any additional supervision other than the raw unconstrained videos themselves, is shown to successfully solve this task, and, more interestingly, result in good vision and audio representations."
-
One Model To Learn Them All (2017) - "We present a single model that yields good results on a number of problems spanning multiple domains. In particular, this single model is trained concurrently on ImageNet, multiple translation tasks, image captioning (COCO dataset), a speech recognition corpus, and an English parsing task."
-
Why does deep and cheap learning work so well? (2016) - "We show how the success of deep learning depends not only on mathematics but also on physics: although well-known mathematical theorems guarantee that neural networks can approximate arbitrary functions well, the class of functions of practical interest can be approximated through "cheap learning" with exponentially fewer parameters than generic ones, because they have simplifying properties tracing back to the laws of physics".
-
One-shot Learning with Memory-Augmented Neural Networks (2016) - "Here, we demonstrate the ability of a memory-augmented neural network to rapidly assimilate new data, and leverage this data to make accurate predictions after only a few samples. We also introduce a new method for accessing an external memory that focuses on memory content, unlike previous methods that additionally use memory location-based focusing mechanisms".
-
Learning to Compose Neural Networks for Question Answering (2016) - "We describe a question answering model that applies to both images and structured knowledge bases. The model uses natural language strings to automatically assemble neural networks from a collection of composable modules. Our approach, which we term a dynamic neural model network, achieves state-of-the-art results on benchmark datasets in both visual and structured domains".
-
A guide to convolution arithmetic for deep learning (2016) - "We introduce a guide to help deep learning practitioners understand and manipulate convolutional neural network architectures".
-
Action-Conditional Video Prediction using Deep Networks in Atari Games (2015) - "Motivated by vision-based reinforcement learning (RL) problems, in particular Atari games from the recent benchmark Aracade Learning Environment (ALE), we consider spatio-temporal prediction problems where future (image-)frames are dependent on control variables or actions as well as previous frames".
-
Understanding Neural Networks Through Deep Visualization (2015) - "Progress in the field will be further accelerated by the development of better tools for visualizing and interpreting neural nets. We introduce two such tools here. The first is a tool that visualizes the activations produced on each layer of a trained convnet as it processes an image or video. (...) The second tool enables visualizing features at each layer of a DNN via regularized optimization in image space".
-
Neural Turing Machines (2014) - "We extend the capabilities of neural networks by coupling them to external memory resources, which they can interact with by attentional processes. The combined system is analogous to a Turing Machine or Von Neumann architecture but is differentiable end-to-end, allowing it to be efficiently trained with gradient descent".
-
Wide & Deep Learning for Recommender Systems - "In this paper, we present Wide & Deep learning - jointly trained wide linear models and deep neural networks - to combine the benefits of memorization and generalization for recommender systems".
- Evolution Strategies as a Scalable Alternative to Reinforcement Learning (2017) - "We explore the use of Evolution Strategies, a class of black box optimization algorithms, as an alternative to popular RL techniques such as Q-learning and Policy Gradients. Experiments on MuJoCo and Atari show that ES is a viable solution strategy that scales extremely well with the number of CPUs available".
-
Properties of Sparse Distributed Representations and their Application to Hierarchical Temporal Memory - "Empirical evidence demonstrates that every region of the neocortex represents information using sparse activity patterns. This paper examines Sparse Distributed Representations (SDRs), the primary information representation strategy in Hierarchical Temporal Memory (HTM) systems and the neocortex".
-
HTM Whitepaper - "Hierarchical Temporal Memory (HTM) is a technology modeled on how the neocortex performs these functions. HTM offers the promise of building machines that approach or exceed human-level performance for many cognitive tasks".
-
Biological and Machine Intelligence (BAMI) - "Biological and Machine Intelligence (BAMI) is a living book authored by Numenta researchers and engineers. Its purpose is to document Hierarchical Temporal Memory, a theoretical framework for both biological and machine intelligence".
-
NIPS 2016 Tutorial: Generative Adversarial Networks (2016) - "This report summarizes the tutorial presented by the author at NIPS 2016 on generative adversarial networks (GANs)".
-
Deep multi-scale video prediction beyond mean square error (2016) - "In this work, we train a convolutional network to generate future frames given an input sequence. To deal with the inherently blurry predictions obtained from the standard Mean Squared Error (MSE) loss function, we propose three different and complementary feature learning strategies: a multi-scale architecture, an adversarial training method, and an image gradient difference loss function".
-
One-Shot Generalization in Deep Generative Models (2016) - "We develop machine learning systems with this important capacity by developing new deep generative models, models that combine the representational power of deep learning with the inferential power of Bayesian reasoning. We develop a class of sequential generative models that are built on the principles of feedback and attention".
-
Improved Techniques for Training GANs (2016) - "We present a variety of new architectural features and training procedures that we apply to the generative adversarial networks (GANs) framework. We focus on two applications of GANs: semi-supervised learning, and the generation of images that humans find visually realistic".
-
Learning What and Where to Draw (2016) - "We propose a new model, the Generative Adversarial What-Where Network (GAWWN), that synthesizes images given instructions describing what content to draw in which location".
-
Generating images with recurrent adversarial networks (2016) - "Gatys et al. (2015) showed that optimizing pixels to match features in a convolutional network with respect reference image features is a way to render images of high visual quality. We show that unrolling this gradient-based optimization yields a recurrent computation that creates images by incrementally adding onto a visual "canvas"".
-
DRAW: A Recurrent Neural Network For Image Generation (2015) - "This paper introduces the Deep Recurrent Attentive Writer (DRAW) neural network architecture for image generation. DRAW networks combine a novel spatial attention mechanism that mimics the foveation of the human eye, with a sequential variational auto-encoding framework that allows for the iterative construction of complex images".
-
Generative Adversarial Networks (2014) - "We propose a new framework for estimating generative models via an adversarial process, in which we simultaneously train two models: a generative model G that captures the data distribution, and a discriminative model D that estimates the probability that a sample came from the training data rather than G".
-
Reading Wikipedia to Answer Open-Domain Questions (2017) - "This paper proposes to tackle open- domain question answering using Wikipedia as the unique knowledge source: the answer to any factoid question is a text span in a Wikipedia article. This task of machine reading at scale combines the challenges of document retrieval (finding the relevant articles) with that of machine comprehension of text (identifying the answer spans from those articles). Our approach combines a search component based on bigram hashing and TF-IDF matching with a multi-layer recurrent neural network model trained to detect answers in Wikipedia paragraphs."
-
Learning to Generate Reviews and Discovering Sentiment (2017) - "We explore the properties of byte-level recurrent language models. When given sufficient amounts of capacity, training data, and compute time, the representations learned by these models include disentangled features corresponding to high-level concepts. Specifically, we find a single unit which performs sentiment analysis."
-
Modeling Human Reading with Neural Attention (2016) - "When humans read text, they fixate some words and skip others. However, there have been few attempts to explain skipping behavior with computational models, as most existing work has focused on predicting reading times (e.g.,~using surprisal). In this paper, we propose a novel approach that models both skipping and reading, using an unsupervised architecture that combines a neural attention with autoencoding, trained on raw text using reinforcement learning."
-
Semi-supervised Sequence Learning (2015) - "We present two approaches that use unlabeled data to improve sequence learning with recurrent networks. The first approach is to predict what comes next in a sequence, which is a conventional language model in natural language processing. The second approach is to use a sequence autoencoder, which reads the input sequence into a vector and predicts the input sequence again. These two algorithms can be used as a "pretraining" step for a later supervised sequence learning algorithm."
-
Character-level Convolutional Networks for Text Classification (2015) - "This article offers an empirical exploration on the use of character-level convolutional networks (ConvNets) for text classification. We constructed several large-scale datasets to show that character-level convolutional networks could achieve state-of-the-art or competitive results. Comparisons are offered against traditional models such as bag of words, n-grams and their TFIDF variants, and deep learning models such as word-based ConvNets and recurrent neural networks."
- Neuroscience-Inspired Artificial Intelligence (2017) - "The fields of neuroscience and artificial intelligence (AI) have a long and intertwined history. In more recent times, however, communication and collaboration between the two fields has become less commonplace. In this article, we argue that better understanding biological brains could play a vital role in building intelligent machines."
-
Curiosity-driven Exploration by Self-supervised Prediction (2017) - "In many real-world scenarios, rewards extrinsic to the agent are extremely sparse, or absent altogether. In such cases, curiosity can serve as an intrinsic reward signal to enable the agent to explore its environment and learn skills that might be useful later in its life. We formulate curiosity as the error in an agent's ability to predict the consequence of its own actions in a visual feature space learned by a self-supervised inverse dynamics model."
-
Human-Level Control through Deep Reinforcement Learning (2015) - "Here we use recent advances in training deep neural networks to develop a novel artificial agent, termed a deep Q-network, that can learn successful policies directly from high-dimensional sensory inputs using end-to-end reinforcement learning. We tested this agent on the challenging domain of classic Atari 2600 games."
-
End to End Learning for Self-Driving Cars (2016) - "We trained a convolutional neural network (CNN) to map raw pixels from a single front-facing camera directly to steering commands".
-
Learning a Driving Simulator (2016) - "Comma.ai's approach to Artificial Intelligence for self-driving cars is based on an agent that learns to clone driver behaviors and plans maneuvers by simulating future events in the road. This paper illustrates one of our research approaches for driving simulation".
-
DeepDriving: Learning Affordance for Direct Perception in Autonomous Driving (2015) - "Today, there are two major paradigms for vision-based autonomous driving systems: mediated perception approaches that parse an entire scene to make a driving decision, and behavior reflex approaches that directly map an input image to a driving action by a regressor. In this paper, we propose a third paradigm: a direct perception approach to estimate the affordance for driving. We propose to map an input image to a small number of key perception indicators that directly relate to the affordance of a road/traffic state for driving".
-
An Empirical Evaluation of Deep Learning on Highway Driving (2015) - "In this paper, we presented a number of empirical evaluations of recent deep learning advances".
-
CS 294: Deep Reinforcement Learning - Deep Reinforcement Learning course by Sergey Levine, John Schulman, Chelsea Finn.
-
Open Source Deep Learning Curriculum - "This open-source Deep Learning curriculum is meant to be a starting point for everyone interested in seriously studying the field".
-
Montreal Institute for Learning Algorithms - MILA research publications.
-
Shakir's Machine Learning Blog - Shakir is a senior research scientist at Google DeepMind in London.
-
A visual proof that neural nets can compute any function - A very nice, simple and mostly visual explanation of the universality theorem, by Michael Nielsen.
-
Building a Deep Learning (Dream) Machine - Guidelines for building a machine system specifically tailored for Deep Learning (2016), by Roelof Pieters.
- Who is the best in dataset X? - A collection of the best performing classification methods in datasets like MNIST and CIFAR-10.
-
Bay Area Deep Learning School - Day 1 (2016) - Day 1 of Bay Area Deep Learning School featuring speakers Hugo Larochelle, Andrej Karpathy, Richard Socher, Sherry Moore, Ruslan Salakhutdinov and Andrew Ng. Detailed schedule is at http://www.bayareadlschool.org/schedule.
-
Bay Area Deep Learning School - Day 2 (2016) - Day 2 of Bay Area Deep Learning School featuring speakers John Schulman, Pascal Lamblin, Adam Coates, Alex Wiltschko, Quoc Le and Yoshua Bengio. Detailed schedule is at http://www.bayareadlschool.org/schedule.