What do Smartphone Predictive Text and Cybersecurity have in common?

Maybe the link between your smartphone keyboard and current machine learning research in cybersecurity is not apparent at first glance, but the technology behind both is extremely similar: both leverage deep learning architectures called Recurrent Neural Networks [RNNs], specifically a type of RNN called Long Short Term Memory [LSTM].

One of the main advantages of LSTMs is their ability to deal with sequences very well. Due to the composition of the building blocks of LSTMs, these RNNs are able to predict the next step in a sequence given previous steps by taking into account not only the statistical properties of a sequence in question (e.g. frequency) but also the temporal properties of a sequence. To give a practical example of “temporal properties”, let’s imagine a simplistic example. Say an LSTM has been trained with sequences similar to the following:

previous steps -> next step

“1 1 1” -> 2

“4 4 4” -> 5

Given the never-before-seen sequence of “8 8 8” the LSTM is very well able to predict “9” correctly. This may seem simplistic but a neural network typically deals with thousands or millions of different sequences, but the LSTM is anyway capable of learning the intuitive rule in our example that if you see three repeated numbers, the next number is simply +1. This is different from spatial or frequency based machine learning techniques (such as One Class SVMs) where a never-before-seen sequence gets classified as an anomaly — precisely because it’s never been seen before.


Your smartphone keyboard is actually powered by deep learning

You probably use LSTMs every day without realizing it — in the form of the predictive text suggestions that appear whenever you are typing something in your smartphone. As we just explained, LSTMs are very good with sequences. Sequences can just as well be letters rather than numbers. So given enough training, given a previous sequence of letters, an LSTM gets very good at suggesting the next letter, couple of letters, or the whole word.

The screenshot above is familiar to all of you… start typing and given a sequence of characters, the LSTM will predict the most probable next few characters. These “predictions” are what we call suggestions.

 


 

Where things get interesting for cybersecurity analysts is what happens when we feed an LSTM a sequence of characters which are abnormal.

 

 

An example of doing this on your smartphone is shown above. When we feed the LSTM an abnormal sequence of characters, it cannot predict with any certainty what the next character is. This manifests itself in very limited suggestions. In the screenshot, note how the keyboard suggestions are limited to the sequence itself (LSTM could not predict the next character, or it simply prepends common characters).

 


The cybersecurity tie-in

One man’s trash is another man’s gold. While the above might not seem very useful to the smartphone user — it is to a cybersecurity analyst who is looking for anomalies within the millions of logs that are generated by security devices.

For example, let’s consider CyberSift’s Docker anomaly detection engine. The concept is pretty simple: detect anomalous sequences of system calls. Any operating system’s activity can be characterized as a stream of system calls like so:

open-read-read-write-poll-listen-accept-close

We can imagine each system call as being a character or number in a longer sequence — exactly what LSTM is designed to handle. To give a practical example, let’s imagine we are using an LSTM that has been trained on common sequences of system calls. Next, we see how the LSTM reacts when we ask it to predict the next system call, given a sequence of syscalls which is relatively common. The LSTM output could look similar to this:

 

 

The above graph shows that the LSTM is 90% certain that the next syscall is going to be “open”. Similar to what we saw before with the smartphone keyboard, the LSTM network has a good chance of being correct.

Contrast this to what happens when we feed the LSTM network an unusual syscall sequence. Just like before, the LSTM network will get confused and give very uncertain predictions:

 

The above graph still shows “open” as being the next most probable system call, but the network is a lot less certain about it (16% vs the 90% we had previously)

 


 

This is exactly how CyberSift leverages deep learning to help detect anomalies in your docker environment — or to detect anomalies within your logs, highlighting those sequences that are different or unusual and therefore are more worthy of your limited time.

These types of protections are becoming increasingly important as novel attacks are discovered against docker and other systems which do not necessarily trigger signatures, but definitely generate anomalous behavior.

 

Source: https://threatpost.com/attack-uses-docker-containers-to-hide-persist-plant-malware/126992/

 

Consider the attack presented in Black Hat just last month — where hackers were able to spin up a docker container just by having a target visiting a specially crafted webpage. Their attack consists of leveraging the docker API to start a container and then use that to laterally attack the network. In a busy docker environment, where containers are being started and stopped multiple times within a short period of time, keeping your eye on all the containers being started may be a bit too much to handle, but as we can see from CyberSift’s anomaly detection engine output below — starting a container that performs unusual actions shows up as a highly anomalous period:

 

Note the significantly higher anomaly score for the time period where a docker container was spun up and performed a range of lateral attacks and data exfiltration. For further information about the test environment used to capture the above results, please have a quick read here

 

For more posts like this, written in my capacity as CTO of CyberSift, please follow us on Medium! We include more technical, marketing, and management articles all relating to InfoSec

https:[email protected]

Advertisements

First Steps in applying machine learning to InfoSec – WEKA

The intersection between machine learning [ML] and information security [InfoSec] is currently quite a hot topic. The allure of this intersection is easy to see, security analysts are drowning in alerts and data which need to be painstakingly investigated and if necessary acted upon. This is no easy processes and as was seen in the now infamous Target hack, more often than not alarms go by unnoticed. ML promises to alleviate the torrent of alerts and logs and (ideally) present to the analyst only those alerts which are really worthwhile investigating.

This is by no means an easy task however the rise of several enabling factors has made this goal reachable to the average InfoSec professional:

  • Cloud Computing
  • Big Data technologies such as Hadoop
  • Python (and other language) libraries like Scikit-Learn [1] which abstract away the nuances of Machine Learning and Data Mining
  • Distributed Data/Log collection and search technologies such as ElasticSearch [2]

From personal experience the process of learning about machine learning can be daunting, especially to those not of a mathematical background. However, in this series of articles I plan on outlining my learning process and enumerating the various excellent resources that are freely available on the internet to help anyone interested in getting started in this exciting field.

A good introduction into this field is a talk by @j_monty and @rsevey about “Using Machine Learning Solutions to Solve Serious Security Problems” which can be found here:

https://youtu.be/48O6L_DfE2o

The talk really whets your appetite for this field. A small distinction that should be pointed out is the difference between “machine learning” and “data mining“. Data mining is the process of turning raw data into actionable information, while machine learning is one of the many tools/algorithms that help in this process. The presenters mention using WEKA [3] to get started in the field and get to grips with understanding the data that will eventually power our algorithms and machine learning. Before anything else, it will be very useful to manually try some data mining techniques to understand our data, which algorithms to apply to this data for best results and understand the challenges and rewards of doing so. This will allow us to better understand which machine learning algorithms we can later apply to infosec related data such as logs, pcaps and so on.

So it would seem WEKA is as good a place as any to get started! Some quick research turns up a hidden gem…. an online course from the creators of WEKA on how to use the program:

https://weka.waikato.ac.nz/dataminingwithweka/preview

The course may not be open when reading this, however the course videos are still available on YouTube and this should be your first stop:

http://www.cs.waikato.ac.nz/ml/weka/mooc/dataminingwithweka/

Note: if you need to find the datasets the instructor is using (the WEKA installation from the Ubuntu repositories do not include these), then you can find them here:

http://storm.cis.fordham.edu/~gweiss/data-mining/datasets.html

References

[1] Scikit-learn : http://scikit-learn.org/stable/

[2] ElasticSearch: https://www.elastic.co/

[3] WEKA: http://www.cs.waikato.ac.nz/ml/weka/