はてなブックマーク

UPDATE 2022: I made this page in 2015. Some of the links below were out of date so I updated them to point to the original 2015 code. This tutorial was made for educational purposes to teach how compilers and libraries obtain high CPU performance. It is not trying to replace BLAS libraries or generate portable code. A common misconception is that BLAS implementations of matrix multiplication are o

I am a research scientist at Meta, building next generation language models, multimodal models, and generative AI. Previously, I completed a PhD in AI at Stanford, advised by Percy Liang, Jure Leskovec and Chris Manning, and worked at Google DeepMind. I am interested in building multimodal foundation models that can assist humans in diverse tasks. In particular, I work on: Multimodal understanding

Redirecting to the latest URL

Toeplitz Inverse Covariance-Based Clustering of Multivariate Time Series Data David Hallac, Sagar Vare, Stephen Boyd, Jure Leskovec Stanford University {hallac,svare,boyd,jure}@stanford.edu ABSTRACT Subsequence clustering of multivariate time series is a useful tool for discovering repeated patterns in temporal data. Once these pat- terns have been discovered, seemingly complicated datasets can be

This page has moved to https://nlp.stanford.edu/projects/jesc/index_ja.html

This page has moved to https://nlp.stanford.edu/projects/jesc/

Visual relationships capture a wide variety of interactions between pairs of objects in images (e.g. "man riding bicycle" and "man pushing bicycle"). Consequently, the set of possible relationships is extremely large and it is difficult to obtain sufficient training examples for all possible relationships. Because of this limitation, previous work on visual relationship detection has concentrated

YellowFin and the Art of Momentum Tuning by Jian Zhang, Ioannis Mitliagkas and Chris Ré. TLDR; Hand-tuned momentum SGD is competitive with state-of-the-art adaptive methods, like Adam. We introduce YellowFin, an automatic tuner for the hyperparameters of momentum SGD. YellowFin trains large ResNets and LSTMs in fewer iterations than the state-of-the-art. It performs even better in asynchronous set

CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning Abstract When building artificial intelligence systems that can reason and answer questions about visual data, we need diagnostic tests to analyze our progress and discover shortcomings. Existing benchmarks for visual question answering can help, but have strong biases that models can exploit to correctly answer

ConvnetJS demo: toy 2d classification with 2-layer neural network The simulation below shows a toy binary problem with a few data points of class 0 (red) and 1 (green). The network is set up as: Feel free to change this, the text area above gets eval()'d when you hit the button and the network gets reloaded. Every 10th of a second, all points are fed to the network multiple times through the train

ConvnetJS demo: Image "Painting" This demo that treats the pixels of an image as a learning problem: it takes the (x,y) position on a grid and learns to predict the color at that point using regression to (r,g,b). It's a bit like compression, since the image information is encoded in the weights of the network, but almost certainly not of practical kind :) Note that the entire ConvNetJS definition

DenseCap: Fully Convolutional Localization Networks for Dense Captioning We introduce the dense captioning task, which requires a computer vision system to both localize and describe salient regions in images in natural language. The dense captioning task generalizes object detection when the descriptions consist of a single word, and Image Captioning when one predicted region covers the full imag

Below every paper are TOP 100 most-occuring words in that paper and their color is based on LDA topic model with k = 7. (This is very hard but it looks like 0 = graphical models?, 1 = reinforcement learning?, 2 = deep learning, 3 = kernels?, 4 = theory?, 5 = optimization, 6 = matrix factorization?)

The simplest way to examine the advantages and disadvantages of RISC architecture is by contrasting it with it's predecessor: CISC (Complex Instruction Set Computers) architecture. Multiplying Two Numbers in Memory On the right is a diagram representing the storage scheme for a generic computer. The main memory is divided into locations numbered from (row) 1: (column) 1 to (row) 6: (column) 4. The

I took 50,000 ILSVRC 2012 validation images, extracted the 4096-dimensional fc7 CNN (Convolutional Neural Network) features using Caffe and then used Barnes-Hut t-SNE to compute a 2-dimensional embedding that respects the high-dimensional (L2) distances. In other words, t-SNE arranges images that have a similar CNN (fc7) code nearby in the embedding.

# About **REINFORCEjs** is a Reinforcement Learning library that implements several common RL algorithms supported with fun web demos, and is currently maintained by [@karpathy](https://twitter.com/karpathy). In particular, the library currently includes: ### Dynamic Programming For solving finite (and not too large), deterministic MDPs. The solver uses standard tabular methods will no bells and w

Deprecated. It's been a while since I graduated from Stanford. My main webpage has moved to karpathy.ai Bio. I am the Sr. Director of AI at Tesla, where I lead the team responsible for all neural networks on the Autopilot. Previously, I was a Research Scientist at OpenAI working on Deep Learning in Computer Vision, Generative Modeling and Reinforcement Learning. I received my PhD from Stanford, wh

Please excuse the retro formatting. This page was created back in early days when raw HTML was the norm, for those of us who actually made web pages back then. I haven't bothered to modernize it, figuring it's the content that actually counts. Here are the notes from a presentation I gave at the Stanford InfoLab Friday lunch, 1/27/06, with a few (not many) revisions when I reprised the talk on 12/

Large-scale Video Classification with Convolutional Neural Networks Abstract Convolutional Neural Networks (CNNs) have been established as a powerful class of models for image recognition problems. Encouraged by these results, we provide an extensive empirical evaluation of CNNs on large-scale video classification using a new dataset of 1 million YouTube videos belonging to 487 classes. We study m

Below every paper are TOP 100 most-occuring words in that paper and their color is based on LDA topic model with k = 7. (It looks like 0 = graphical learning?, 1 = reinforcement learning, 2 = deep learning, 3 = non-parametrics?, 4 = matrix factorization?, 5 = neuroscience, 6 = optimization)