Lab41

31 Flips | 1 Magazine | 1 Like | 3 Followers | @Lab41 | Keep up with Lab41 on Flipboard, a place to see the stories, photos, and updates that matter to you. Flipboard creates a personalized magazine full of everything, from world news to life’s great moments. Download Flipboard for free and search for “Lab41”

Iterating through Siamese network designs

<b>TL;DR — first, iterate quickly with a time memory trade off, then rank architecture performance, finally refine by rank order.</b><p>I recently finished …

Deep Session Learning for Cyber Security

Recently, Lab41 teamed up with Cyber Reboot (a sister lab) to explore the intersection of deep learning (DL) and cyber security in a software defined …

Machine Learning

TL;DR — pre-commit hooks can save you from embarrassment and improve workflow

TL;DR — pre-commit hooks can save you from embarrassment and improve workflow<p>Workflow Improvements<p>Not everything is big data, machine or deep …

Python Programming

Sampled Backpropagation

Training Deep and Wide Neural Networks on Large Scale, User Generated Content Using Label Sampling<p>You might have heard of this thing called deep …

Deep Learning

Docker, the great equalizer

Docker allows you to describe a consistent, reproducible container for your executable. The container provides lightweight isolation of the …

DevOps

Speech Recognition: You down with CTC?

Demystifying the Connectionist Temporal Classification Loss<p>You may have heard that speech recognition nowadays does away with everything that’s not a …

Deep Learning

Mistakes I’ve Made in Angular 1

And is it right to choose Angular as the front-end?<p>As excited as we are to dive right in to the heart of the Gestalt challenge — to explain data …

JavaScript

Across the Network — AI Week in Review Sept 23

Welcome back to Across the Network — Lab41’s weekly look at what is going on in the world of AI. As always these are all links that I pulled from the …

Redirecting…

Tell me something I don’t know: Detecting novelty and redundancy with natural language processing

Introducing Pythia, a new Lab41 challenge<p>Large collections of documents can be hard for humans to sift through on their own. High-quality search can …

Is Optimizing your Neural Network a Dark Art ?

Just like in real life, different people learn at different pace, using different techniques and understand or retain qualitatively different aspects …

More

1.1 Billion Taxi Rides with MapD & 8 Nvidia Pascal Titan Xs

On August 2nd, Nvidia released a new Pascal-based Titan X graphics card. This new card brought an architectural upgrade over the previous …

Build Algorithms Like You Give a Damn

<i>Source: Sam Charrington via Twitter</i><p>For the second year in a row, WrangleConf did not disappoint. The conversation picked up right where last year’s …

Personal Information

AI’s Language Problem - MIT Technology Review

About halfway through a particularly tense game of Go held in Seoul, South Korea, between Lee Sedol, one of the best players of all time, and …

The Ultimate List of TensorFlow Resources

A curated list of 50+ awesome TensorFlow resources including tutorials, books, libraries, projects and more.<p>If you know of any awesome TensorFlow …

Deep Compression

The first compression method is <b>Network Pruning</b>. In this method a network is fully trained and then any connections with a weight below a certain …

Mind-blowing app Prisma makes Instagram’s filters look lame

There was a time when people thought a painter’s work was so intricate and personal, it was deemed impossible to recreate.<p>Well, that time is over. With Prisma, you now have your own, personal Picasso in your pocket. Just pick a photo from your phone, and it starts painting.<p>By using a combination of …

Apps

Deep Learning and the Future of AI

Speaker: Prof. LeCun Yann

Deep Learning

Machine learning for artists

This spring I will be teaching a course at NYU’s Interactive Telecommunications Program (ITP) called “Machine Learning for Artists.” Since the …

Batch Normalization — What the hey?

A little while ago, you might have read about batch normalization being the next coolest thing since ReLu’s. Things have since moved on, but it’s …

Deep Learning

Fits like a GloVe: “Betta” style with dense word vectors

Word embeddings — think word2vec, GloVe and their many progeny — are a representation of the relationships between words, and we have obsessed about …

Natural Language Processing

Taking Keras to the Zoo

If you follow any of the popular blogs like Google’s research, FastML, Smola’s Adventures in Data Land, or one of the indie-pop ones like Edwin …

Deep Learning

How predictive analytics discovers a data breach before it happens

Ben Dickson Contributor<p>Ben Dickson is a software engineer and the founder of TechTalks.<p>More posts by this contributor:<p>Unlocking the potential of eye tracking technology<br>• Can you trust crypto-token crowdfunding?<p>Cybersecurity experts and analysts are constantly trying to keep pace with changes and …

Anything2Vec

2Vec or Not 2Vec?<p>This might be old news to you, but if you’re considering the use of word embeddings, our suggestion: just take the plunge. We’ve …

Deep Learning

Commit-and-Push to GitHub from Jupyter Notebooks

In our post on deep learning environments, we described how we’re using Jupyter-driven Docker containers for a faster on-ramp and better …

DevOps

Tales from ICML

This year I had the opportunity to attend the International Conference for Machine Learning (ICML) in New York City, and overall, I quite enjoyed the …

Machine Learning

Five Things I Learned at CVPR 2016

“Understanding” is the <i>New Hotness</i><p>Since this is a wrap-up post, it makes sense to try and describe the current “hot topics” and trends. An easy way to …

Deep Learning

I need an AI BS-Meter

We talk to a lot of analysts at Lab41. ​A recurring theme of these conversations is what they frequently refer to as “result provenance.” …

Machine Learning

Tensorflow 3 Ways

I recently started my first real project in Tensorflow. The larger effort, Attalos, was aimed at exploring vector spaces containing multiple types …