Google Research Blog
The latest news from Research at Google
Distill: Supporting Clarity in Machine Learning
Monday, March 20, 2017
Posted by Shan Carter, Software Engineer and Chris Olah, Research Scientist, Google Brain Team
Science isn't just about discovering new results. It’s also about human understanding. Scientists need to develop notations, analogies, visualizations, and explanations of ideas. This human dimension of science isn't a minor side project. It's deeply tied to the heart of science.
That’s why, in collaboration with OpenAI, DeepMind, YC Research, and others, we’re excited to announce the launch of
Distill
, a new open science journal and ecosystem supporting human understanding of machine learning. Distill is an independent organization, dedicated to fostering a new segment of the research community.
Modern web technology gives us powerful new tools for expressing this human dimension of science. We can create interactive diagrams and user interfaces the enable intuitive exploration of research ideas. Over the last few years we've seen
many
incredible
demonstrations
of
this
kind
of
work
.
An interactive diagram explaining the Neural Turing Machine from
Olah & Carter, 2016
.
Unfortunately, while there are a plethora of conferences and journals in machine learning, there aren’t any research venues that are dedicated to publishing this kind of work. This is partly an issue of focus, and partly because traditional publication venues can't, by virtue of their medium, support interactive visualizations. Without a venue to publish in, many significant contributions don’t count as “real academic contributions” and their authors can’t access the academic support structure.
That’s why Distill aims to build an ecosystem to support this kind of work, starting with three pieces: a research journal, prizes recognizing outstanding work, and tools to facilitate the creation of interactive articles.
Distill is an ecosystem to support clarity in Machine Learning
.
Led by a diverse steering committee of leaders from the machine learning and user interface communities, we are very excited to see where Distill will go. To learn more about Distill, see the
overview page
or read the
latest articles
.
Open sourcing the Embedding Projector: a tool for visualizing high dimensional data
Wednesday, December 07, 2016
Posted by Daniel Smilkov and the Big Picture group
Recent advances in Machine Learning (ML) have shown impressive results, with applications ranging from
image recognition
,
language translation
,
medical diagnosis
and more. With the widespread adoption of ML systems, it is increasingly important for research scientists to be able to explore how the data is being interpreted by the models. However, one of the main challenges in exploring this data is that it often has hundreds or even thousands of dimensions, requiring special tools to investigate the space.
To enable a more intuitive exploration process, we are
open-sourcing the Embedding Projector
, a web application for interactive visualization and analysis of high-dimensional data recently shown as an
A.I. Experiment
, as part of
TensorFlow
. We are also releasing a standalone version at
projector.tensorflow.org
, where users can visualize their high-dimensional data without the need to install and run TensorFlow.
Exploring Embeddings
The data needed to train machine learning systems comes in a form that computers don't immediately understand. To translate the things we understand naturally (e.g. words, sounds, or videos) to a form that the algorithms can process, we use
embeddings
, a mathematical vector representation that captures different facets (dimensions) of the data. For example, in
this language embedding
, similar words are mapped to points that are close to each other.
With the Embedding Projector, you can navigate through views of data in either a 2D or a 3D mode, zooming, rotating, and panning using natural click-and-drag gestures. Below is a figure showing the nearest points to the embedding for the word “important” after training a TensorFlow model using the
word2vec tutorial
. Clicking on any point (which represents the learned embedding for a given word) in this visualization, brings up a list of nearest points and distances, which shows which words the algorithm has learned to be semantically related. This type of interaction represents an important way in which one can explore how an algorithm is performing.
Methods of Dimensionality Reduction
The Embedding Projector offers three commonly used methods of data dimensionality reduction, which allow easier visualization of complex data:
PCA
,
t-SNE
and custom linear projections.
PCA
is often effective at exploring the internal structure of the embeddings, revealing the most influential dimensions in the data.
t-SNE
, on the other hand, is useful for exploring local neighborhoods and finding clusters, allowing developers to make sure that an embedding preserves the meaning in the data (e.g. in the
MNIST dataset
, seeing that the same digits are clustered together). Finally, custom linear projections can help discover meaningful "directions" in data sets - such as the distinction between a formal and casual tone in a language generation model - which would allow the design of more adaptable ML systems.
A custom linear projection of the 100 nearest points of "See attachments." onto the "yes" - "yeah" vector (“yes” is right, “yeah” is left) of a corpus of
35k frequently used phrases in emails
The Embedding Projector
website
includes a few datasets to play with. We’ve also made it easy for users to publish and share their embeddings with others (just click on the “Publish” button on the left pane). It is our hope that the
Embedding Projector
will be a useful tool to help the research community explore and refine their ML applications, as well as enable anyone to better understand how ML algorithms interpret data. If you'd like to get the full details on the Embedding Projector, you can read the paper
here
. Have fun exploring the world of embeddings!
Open Source Visualization of GPS Displacements for Earthquake Cycle Physics
Thursday, November 10, 2016
Posted by Jimbo Wilson, Software Engineer, Google Big Picture Team and Brendan Meade, Professor, Harvard Department of Earth and Planetary Sciences
The
Earth’s surface is moving
, ever so slightly, all the time. This slow, small, but persistent movement of the Earth's crust is responsible for the formation of mountain ranges, sudden earthquakes, and even the positions of the continents. Scientists around the world measure these almost imperceptible movements using arrays of
Global Navigation Satellite System
(GNSS) receivers to better understand all phases of an earthquake cycle—both how the surface responds after an earthquake, and the storage of
strain energy
between earthquakes.
To help researchers explore this data and better understand the Earthquake cycle, we are releasing a new, interactive data visualization which draws geodetic velocity lines on top of a relief map by amplifying position estimates relative to their true positions. Unlike existing approaches, which focus on small time slices or individual stations, our visualization can show all the data for a whole array of stations at once. Open sourced under an
Apache 2 license
, and
available on GitHub
, this visualization technique is a collaboration between Harvard’s
Department of Earth and Planetary Sciences
and Google's
Machine Perception
and
Big Picture
teams.
Our approach helps scientists quickly assess deformations across all phases of the earthquake cycle—both during earthquakes (coseismic) and the time between (interseismic). For example, we can see azimuth (direction) reversals of stations as they relate to topographic structures and active faults. Digging into these movements will help scientists vet their models and their data, both of which are crucial for developing accurate computer representations that may help predict future earthquakes.
Classical approaches to visualizing these data have fallen into two general categories: 1) a map view of velocity/displacement vectors over a fixed time interval and 2) time versus position plots of each GNSS component (longitude, latitude and altitude).
Examples of classical approaches. On the left is a map view showing average velocity vectors over the period from 1997 to 2001[1]. On the right you can see a time versus eastward (longitudinal) position plot for a single station.
Each of these approaches have proved to be informative ways to understand the spatial distribution of crustal movements and the time evolution of solid earth deformation. However, because geodetic shifts happen in almost imperceptible distances (mm) and over long timescales, both approaches can only show a small subset of the data at any time—a condensed average velocity per station, or a detailed view of a single station, respectively. Our visualization enables a scientist to see all the data at once, then interactively drill down to a specific subset of interest.
Our visualization approach is straightforward; by magnifying the daily longitude and latitude position changes, we show tracks of the evolution of the position of each station. These magnified position tracks are shown as trails on top of a shaded relief topography to provide a sense of position evolution in geographic context.
To see how it works in practice, let’s step through an an example. Consider this tiny set of longitude/latitude pairs for a single GNSS station, with the differing digits shown in bold:
Day Index
Longitude
Latitude
0
139.069904
07
34.949757
897
1
139.069904
00
34.949757
882
2
139.069904
13
34.949757
941
3
139.069904
09
34.949757
921
4
139.069904
13
34.949757
904
If we were to draw line segments between these points directly on a map, they’d be much too small to see at any reasonable scale. So we take these minute differences and multiply them by a user-controlled scaling factor. By default this factor is 10
5.5
(about 316,000x).
To help the user identify which end is the start of the line, we give the start and end points different colors and interpolate between them. Blue and red are the default colors, but they’re user-configurable. Although day-to-day movement of stations may seem erratic, by using this method, one can make out a general trend in the relative motion of a station.
Close-up of a single station’s movement during the three year period from 2003 to 2006.
However, static renderings of this sort suffer from the same problem that velocity vector images do; in regions with a high density of GNSS stations, tracks overlap significantly with one another, obscuring details. To solve this problem, our visualization lets the user interactively control the time range of interest, the amount of amplification and other settings. In addition, by animating the lines from start to finish, the user gets a real sense of motion that’s difficult to achieve in a static image.
We’ve applied our new visualization to the ~20 years of data from the
GEONET array in Japan
. Through it, we can see small but coherent changes in direction before and after the great 2011 Tohoku earthquake.
GPS data sets (in .json format) for both the GEONET data in Japan and the Plate Boundary Observatory (PBO) data in the western US are available at
earthquake.rc.fas.harvard.edu
.
This short animation shows many of the visualization’s interactive features. In order:
Modifying the multiplier adjusts how significantly the movements are magnified.
We can adjust the time slider nubs to select a particular time range of interest.
Using the map controls provided by the
Google Maps JavaScript API
, we can zoom into a tiny region of the map.
By enabling map markers, we can see information about individual GNSS stations.
By focusing on a stations of interest, we can even see curvature changes in the time periods before and after the event.
Station designated 960601 of Japan’s GEONET array is located on the island of Mikura-jima. Here we see the period from 2006 to 2012, with movement magnified 10
5.1
times (126,000x).
To achieve fast rendering of the line segments, we created a custom overlay using
THREE.js
to render the lines in WebGL. Data for the GNSS stations is passed to the GPU in a data texture, which allows our vertex shader to position each point on-screen dynamically based on user settings and animation.
We’re excited to continue this productive collaboration between Harvard and Google as we explore opportunities for groundbreaking, new earthquake visualizations. If you’d like to try out the visualization yourself, follow the instructions at
earthquake.rc.fas.harvard.edu
. It will walk you through the setup steps, including how to download the available data sets. If you’d like to report issues, great! Please submit them through the GitHub project page.
Acknowledgments
We wish to thank Bill Freeman, a researcher on
Machine Perception
, who hatched the idea and developed the initial prototypes, and Fernanda Viégas and Martin Wattenberg of the
Big Picture Team
for their visualization design guidance.
References
[1] Loveless, J. P., and Meade, B. J. (2010).
Geodetic imaging of plate motions, slip rates, and partitioning of deformation in Japan
,
Journal of Geophysical Research.
Google Databoard: A new way to explore industry research
Tuesday, July 09, 2013
Posted by Adam Grunewald, Mobile Marketing Manager
It’s important for people to stay up to date about the most recent research and insights related to their work or personal lives. But it can be difficult to keep up with all the new studies and updated data that’s out there. To make life a bit easier, we’re introducing a new take on how research can be presented. The
Databoard for Research Insights
enables people to explore and interact with some of Google’s recent research in a unique and immersive way. The Databoard uses responsive design to to offer an engaging experience across devices. Additionally, the tool is a new venture into data visualization and shareability with bite-sized charts and stats that can be shared with your friends or coworkers. The Databoard is currently home to several of Google’s market research studies for businesses, but we believe that this way of conveying data can work across all forms of research.
Here are some of the things that make the Databoard different from other ways research is released today:
Easy to use
All of the information in the Databoard is presented in a bite-sized way so that you can quickly find relevant information. You can explore an entire study or jump straight to the topics or data points you care about. The Databoard is also optimized for all devices so you can explore the research on your computer, tablet or smartphone.
Meant to be shared
Most people, when they find a compelling piece of data, want to share it! Whether it’s with a colleague, client, or a community on a blog or social network, compelling insights and data are meant to be shared. With the databoard, you can easily share individual charts and insights or collections of data with anyone through email or social networks, just look for the share button at the top of each chart or insight.
Create a cohesive story
Most research studies set out to answer a specific question, like how people use their smartphones in stores, or how a specific type of consumer shops. This means that businesses need to look across multiple pieces of research to craft a comprehensive business or marketing strategy. With this in mind, the Databoard lets you curate a customized infographic out of the charts or data points you find important across multiple Google research studies. Creating an infographic is quick and easy, and you can share the finished product with your friends or colleagues.
The databoard is currently home to six research studies including
The New Multi-screen World
,
Mobile In-store shopper research
and
Mobile search moments
. New studies will be added frequently. To get started creating your own infographic,
visit the Databoard now
.
Big Pictures with Big Messages
Thursday, July 26, 2012
Posted by Maggie Johnson, Director of Education and University Relations
Google’s Eighth Annual
Computer Science Faculty Summit
opened today in Mountain View with a fascinating talk by Fernanda Viégas and Martin Wattenberg, leaders of the data visualization group at our Cambridge office. They provided insight into their design process in visualizing big data, by highlighting Google+ Ripples and a map of the wind they created.
To preface his explanation of the design process, Martin shared that his team “wants visualization to be ‘G-rated,’ showing the full detail of the data - there’s no need to simplify it, if complexity is done right.” Martin discussed how their
wind map
started as a personal art project, but has gained interest particularly among groups that are interested in information on the wind (sailors, surfers, firefighters). The map displays surface wind data from the
US National Digital Forecast Database
and updates hourly. You can zoom around the United States looking for where the winds are fastest - often around lakes or just offshore - or check out the
gallery
to see snapshots of the wind from days past.
Fernanda discussed the development of
Google+ Ripples
, a visualization that shows
how news spreads
on Google+. The visualization shows spheres of influence and different patterns of spread. For example, someone might post a video to their Google+ page and if it goes viral, we’ll see several circles in the visualization. This depicts the influence of different individuals sharing content, both in terms of the number of their followers and the re-shares of the video, and has revealed that individuals are at times more influential than organizations in the social media domain.
Martin and Fernanda closed with two important lessons in data visualization: first, don’t “dumb down” the data. If complexity is handled correctly and in interesting ways, our users find the details appealing and find their own ways to interact with and expand upon the data. Second, users like to see their personal world in a visualization. Being able to see the spread of a Google+ post, or zoom in to see the wind around one’s town is what makes a visualization personal and compelling-- we call this the “I can see my house from here” feature.
The
Faculty Summit
will continue through Friday, July 27 with talks by Googlers and faculty guests as well as breakout sessions on specific topics related to this year’s theme of digital interactions. We will be looking closely at how computation and bits have permeated our everyday experiences via smart phones, wearable computing, social interactions, and education.
We will be posting here throughout the summit with updates and news as it happens.
Labels
accessibility
ACL
ACM
Acoustic Modeling
Adaptive Data Analysis
ads
adsense
adwords
Africa
AI
Algorithms
Android
Android Wear
API
App Engine
App Inventor
April Fools
Art
Audio
Australia
Automatic Speech Recognition
Awards
Cantonese
Chemistry
China
Chrome
Cloud Computing
Collaboration
Computational Imaging
Computational Photography
Computer Science
Computer Vision
conference
conferences
Conservation
correlate
Course Builder
crowd-sourcing
CVPR
Data Center
Data Discovery
data science
datasets
Deep Learning
DeepDream
DeepMind
distributed systems
Diversity
Earth Engine
economics
Education
Electronic Commerce and Algorithms
electronics
EMEA
EMNLP
Encryption
entities
Entity Salience
Environment
Europe
Exacycle
Expander
Faculty Institute
Faculty Summit
Flu Trends
Fusion Tables
gamification
Gmail
Google Books
Google Brain
Google Cloud Platform
Google Docs
Google Drive
Google Genomics
Google Maps
Google Photos
Google Play Apps
Google Science Fair
Google Sheets
Google Translate
Google Trips
Google Voice Search
Google+
Government
grants
Graph
Graph Mining
Hardware
HCI
Health
High Dynamic Range Imaging
ICLR
ICML
ICSE
Image Annotation
Image Classification
Image Processing
Inbox
Information Retrieval
internationalization
Internet of Things
Interspeech
IPython
Journalism
jsm
jsm2011
K-12
KDD
Klingon
Korean
Labs
Linear Optimization
localization
Low-Light Photography
Machine Hearing
Machine Intelligence
Machine Learning
Machine Perception
Machine Translation
Magenta
MapReduce
market algorithms
Market Research
Mixed Reality
ML
MOOC
Moore's Law
Multimodal Learning
NAACL
Natural Language Processing
Natural Language Understanding
Network Management
Networks
Neural Networks
Nexus
Ngram
NIPS
NLP
On-device Learning
open source
operating systems
Optical Character Recognition
optimization
osdi
osdi10
patents
ph.d. fellowship
PhD Fellowship
PhotoScan
PiLab
Pixel
Policy
Professional Development
Proposals
Public Data Explorer
publication
Publications
Quantum Computing
renewable energy
Research
Research Awards
resource optimization
Robotics
schema.org
Search
search ads
Security and Privacy
Semi-supervised Learning
SIGCOMM
SIGMOD
Site Reliability Engineering
Social Networks
Software
Speech
Speech Recognition
statistics
Structured Data
Style Transfer
Supervised Learning
Systems
TensorFlow
TPU
Translate
trends
TTS
TV
UI
University Relations
UNIX
User Experience
video
Video Analysis
Virtual Reality
Vision Research
Visiting Faculty
Visualization
VLDB
Voice Search
Wiki
wikipedia
WWW
YouTube
Archive
2017
May
Apr
Mar
Feb
Jan
2016
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2015
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2014
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2013
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2012
Dec
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2011
Dec
Nov
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2010
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2009
Dec
Nov
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2008
Dec
Nov
Oct
Sep
Jul
May
Apr
Mar
Feb
2007
Oct
Sep
Aug
Jul
Jun
Feb
2006
Dec
Nov
Sep
Aug
Jul
Jun
Apr
Mar
Feb
Feed
Google
on
Follow @googleresearch
Give us feedback in our
Product Forums
.