pilooch

See https://github.com/beniz/deepdetect/issues/126#issuecomment and other links online. The trick is to install CUDA 8 without the attached driver, then install the new NVidia driver. Once you got that, install on 16.04 is straightforward.

Wrong stance on SSD: always get an SSD where to put the temporary data, i.e. the form that your dataset takes to be fed into the DL backend. For instance, working with Caffe, your image dataset is turned into an LMDB or HDF5 database, while you can keep your unprocessed dataset on a regular HDD. The trick is to always prevent your GPU(s) from starving.

Google wants car-bots to roam the streets and crawl the real world, much as they did in the digital space. They just need a flow of automated cars to blend into.

The video was removed! Anyone with a copy ?

Weird, my mailboxes each contain 10K+ emails and load in a snap.

Well, check this one out: https://www.runabove.com/titan-x-gpu-servers.xml

Ah, it was when it got posted to HN! Count was ~9K.

> Also, DeepDetect! I keep trying to find that and never can remember the name.

Ah, thanks for the reference, DD author here, the funny fact about the name is that translated to my mother tongue (French), it sounds like porn ;)

Nop, check the stats of Caffe for instance: https://github.com/BVLC/caffe

This is much older than the 9th of August.

This list is old :)

We're talking hundreds of thousands of malwares here.

Hey, just finished a malware ML custom system for one of the largest european corporations, large enough that some malware is targeted at them. Result is 97% accuracy (they did retrain and check on their own held out dataset). More careful analysis is needed (many malware have high entropy 'zones' that may help the classifier find the right category), but overall it does work.

See the Microsoft / Kaggle challenge on classifying malware families, winning solution is > 99% accuracy IIRC.

AFAIK not really. Detecting C&C is not that easy in unsupervised settings. The MIT paper (link in the TC article) makes claims, but this is mostly PR.

There are some reports from Google as well that you may find around, and at least at the time (few years back) they make the same observation we do with my coworkers: the rate of false positive remains high, and thus it is the job of trained security operators to spot the sophisticated attacks.

Empowering the operators with more data mining tools is a good step not yet fulfilled I believe.

Machine learning is pervasive in the new products targeted at cybersecurity and infosec. From malware to anomaly detection among billions of logs, it's been part of the demand of my customers for at least four years now. Most of the largest companies are already running and testing a series of novel proprietary or custom in-house systems.

The main differences with straight 'supervised' machine learning are the lack of labels, and the unreal volume of data (cheaply produced by machines, in machine time, we're talking microseconds to milliseconds here). So unsupervised learning is king in this domain, and often security operators have to keep an eye on and interpret the results. Another difference with other fields is that datasets are rare, mostly because of privacy as the logs can be very revealing. For this reason the market exhibits a lock that cannot be overcome by everyone I believe. Basically, you sort of need to be in the place already.

Here is a recent very high level survey on the topic (a TC report, not that bad for once), https://techcrunch.com/2016/07/01/exploiting-machine-learnin...

Here are two of our own Open Source tooling we use the most in application:

- very efficient C++ map-reduce feature generation for logs (for ML and analytics): https://github.com/soprasteria/cybersecurity-miw

- machine learning / deep learning server: https://github.com/beniz/deepdetect

The ML cybersecurity + infosec field is still young, but moving very fast, a lot of new startups and (somewhat opaque) products.

Commoditizing deep learning is mandatory. After repetitive in production installs at various corps while connecting to existing pipelines, I ve convinced some of them to sponsor a commoditized open source deep learning server.

Code is here: https://github.com/beniz/deepdetect

There are differenciated CPU and GPU docker versions, and as mentioned elsewhere in this thread, they are the easiest way to setup even production system without critical impact on performances, thanks to nvidia-docker. It seems they are more popular than AMI within our little community.

Would it work for whistleblowers ? Using some algorithm to cover the voice, could it be a great dumping place for a lot of short or long confessions, advice's, leaks etc.. ? loving the idea already...

> much smaller than a website

The character-based net reads the URL, not the website. The common alphabet used in the model made public is of size 69, URL length can be for example 256.

Not only this is below the image size you are talking about, but after the first convolution the alphabet (in one-hot vector encoding) is collapsed, which leaves convolutions across length 256 in 1D.

Not that terrible. In proxy-like territory, passing batches of URLs to the GPUs could pave the way to new ad-blockers at large. To be tested :)

About the author, https://engineering.purdue.edu/gekcogrp/research-group/JeanM...

Possibly. Having millions of URLs to populate each class is a good thing to start with. Gathered through other means, our current dataset has around 10M URLs in the 'ads' category. The model we made available to the public was built from 2M of these URLs.

EDIT: of possible interest is that these models output a probability and possibly a confidence of having a URL blocked. Base on these, an blocker could ask for confirmation.

Performance is in ms, see http://caffe.berkeleyvision.org/performance_hardware.html Lookup the testing numbers, 500 images / sec. Character-based models are way simpler, you can use test very easily using the original link to the model.

They are not easily gamed either: they learn features as convolutional filters over the characters. These filters are more powerful than ngrams, as complex as the filters you may have seen for images.

Now, about gaming the nets, there's an issue, which is the same as for image-based CNNs: some combinations of letters (or pixel) do exist that do not make a difference to the eye, but push the classifier into the wrong class. However, in order to find these combinations, having a first-hand on the underlying neural net and its weights is mandatory.

Not inside a browser of course. If you consider enterprise-wide applications, NN would be better and possibly faster (with GPU) than a bluecoat proxy.

Since deep learning is dominating the news these days, this might be the opportunity to point to the potential new ways of dealing with ad blocking. One is using character-based convolutional (or recurrent) networks that read the URL characters and classify it as legit or ad.

The interesting gain here is that there's no more conflicting regex nor optimization to work on huge lists of blocking rules. Instead, URLs can be passed to the network in batches.

I have conducted my own set of experiments and a usable deep model is available (see http://www.deepdetect.com/applications/text_model/ ).

This is kinda shameful plug, but I'm really interested in feedback on these new ways of ad blocking with high accuracy.

For sure running a deep net service is not as easy as installing ublock but there are ways. The whole source code is Open Source, as is the model. I have more data at hand, and some larger models could be built. Performance assessment would be a good next step as well.

EDIT: the linked page is rather long, lookup 'Novel task' to get to a quick classification example

An implementation of image similarity search based on deep convnets can be found at https://github.com/beniz/deepdetect/tree/master/demo/imgsear...

Once you're on to training and not using off-the-shelf models, there are several good open source packages out there, from digits to deepdetect. It is not that difficult anymore to get at least a first model with decent accuracy.

You don't need a company for building a meta-search engine. And most importantly, if you value privacy, you want the whole meta-thing to be open source, always.

DDG is basically a meta-search engine. It is then better to host it yourself with https://github.com/asciimoo/searx

Set it up remotely and you basicially have your own DDG.

One of the various open source alternatives, http://www.deepdetect.com/applications/model/

link to your project ?

Google is certainly Nvidia's top customer.

Well, wait to see how support goes etc... In the meantime take on Keras or Lasagne, and do your stuff, i.e. acquire hands-on experience now. When time comes that you need to go into production, or you need a model that is not available in your current lib or framework, then consider a switch. Point being that experience acquired with one framework will empower you whatsover.

In my view, these frameworks are like programming languages, you do need to pick a new one up from time to time, but you can do most of the things with all of them.