The realities of AI in cybersecurity: catastrophic forgetting

Issue 2 2021 Cyber Security

There is a lot of hype about the use of artificial intelligence (AI) in cybersecurity. The truth is that the role and potential of AI in security is still evolving and often requires experimentation and evaluation.

Malware detection is the cornerstone of IT security and AI is the only approach capable of learning patterns from millions of new malware samples within a matter of days.

But there’s a catch: should the model keep all malware samples forever for optimum detection but slower learning and updates; or go for selective fine-tuning that enables the model to better keep up with the rate of change of malware, but runs the risk of forgetting older patterns (known as catastrophic forgetting)?

Retraining the whole model takes about one week. A good fine-tuning model should take about one hour to update.

SophosAI wanted to see if it was possible to have a fine-tuning model that could keep up with the evolving threat landscape, learn new patterns but still remember older ones, while minimising the impact on performance. Researcher Hillary Sanders evaluated a number of update options and has detailed her findings in the Sophos AI blog (https://ai.sophos.com/blog/).

The detection dilemma

Keeping detection capabilities up to date is a constant battle. With every step we take towards defending against a malicious attack, adversaries are already developing new ways to get round it, releasing updates with different code or techniques. The result is that hundreds of thousands of new malware samples appear every day.

Detection is made even harder by the fact that the latest-and-greatest malware is rarely completely ‘new’. Instead, it is more likely to be a combination of new, old, shared, borrowed or stolen code and adopted and adapted behaviours. Further, old malware can re-emerge after years in the wilderness, co-opted into an adversary’s latest arsenal to take defences by surprise.

Detection models need to ensure they can continue to detect older malware samples and not just the most recent ones.

Updating AI detection models

When it comes to updating AI detection models with new malware samples, vendors have a choice between two options.

The first is to keep a copy of every sample they might ever want to detect and retrain the model repeatedly on an ever-increasing volume of data. This results in better overall performance, but also slower updates and fewer releases.

The second is to only update the detection model on new samples. This is known as fine-tuning. During each step of the fine-tuning process, the model updates its understanding according to the new knowledge added and the impact of this on the patterns seen overall. As a result, the model can ‘forget’ the old patterns it learned previously (catastrophic forgetting). However, training a model on less data means the model updates faster and can be released more frequently, keeping better pace with the rapid rate of change of malware.

Regardless of the option chosen, the need to keep training AI detection models on new samples is critical.

The patterns that AI learns from malware samples enable it to generalise and detect not only what it was trained on, but also never before seen samples that bear at least some resemblance to the training data. Over time, however, new samples will begin to deviate enough that an old model’s effectiveness will decay and it will need to be updated.

The three detection update options evaluated by Hillary Sanders were:

1. Learning based on a selection of old and new samples

This is called ‘data-rehearsal’ and involves taking a small selection of old samples and mixing them in with the new, never-before-seen training data. Using this, the model is ‘reminded’ of the old information it needed to detect older samples, while at the same time learning to detect the newer ones.

2. Learning rate

This approach involves modifying how quickly the model learns by adjusting how much it can change after seeing any given sample. If the learning rate is too fast (in which case the model can change a lot with each sample added), it will only remember the most recent samples that it has seen. If the learning rate is too slow (the model can change only slightly with each sample added), it takes too long to learn anything. Finding the right trade-off between learning rate, retaining old information and adding new information can be tricky.

3. Elastic Weight Consolidation (EWC)

This approach was inspired by work by Google’s DeepMind in 2017 and it involves using the old model like an elastic spring to ‘pull back’ the new model if it starts to forget. For a more in-depth explanation of how to implement this approach, read Hillary Sanders’ blog post at https://ai.sophos.com/2021/02/02/catastrophic-forgetting-part-1/.

Findings

All three approaches performed better on older malware samples than on newer samples. Both the EWC and learning-rate approaches remove the need and cost of maintaining older data. However, the graph shows that while their future performance (using new data) is stronger than that achieved using the data-rehearsal technique, they don’t perform as well as data-rehearsal when it comes to remembering past data.

Because the data-rehearsal technique enables faster training and update releases, dips in future performance are more short term and therefore less worrying. Overall, the research showed that the data-rehearsal approach offers the best compromise between simplicity, update speed and performance in malware detection modelling.

Conclusion

In the malware detection game, being able to remember the past is almost as important as being able to predict the future. This must be balanced against the cost and speed of updating your model with new information. Data-rehearsal is a simple and effective way to protect the model’s ability to detect old malware while significantly increasing the pace at which you can update and release new models.

Read more at https://ai.sophos.com/




Share this article:
Share via emailShare via LinkedInPrint this page



Further reading:

Dahua’s cybersecurity approach
CCTV Handbook 2021, Dahua Technology South Africa , Cyber Security
The world is getting smarter and everything is going to have an online ID and then connect into a vast network of IoT devices, like a laptop computer, a mobile phone, a connected thermostat and more.

Read more...
Dahua Technology’s cybersecurity approach
Issue 3 2021, Dahua Technology South Africa , CCTV, Surveillance & Remote Monitoring, Cyber Security
With a mindset that emphasises cybersecurity and all the resources it can allocate to establish, carry out and strengthen its cybersecurity approach, Dahua plans to stay positive, open, responsible and constantly improving its cybersecurity.

Read more...
SASE-enriched threat protection
Issue 3 2021 , Cyber Security
MVISION XDR automates security investigation and response processes with actionable threat insights harnessed from deeply integrated cloud data sources, with the ability to proactively stop targeted attacks.

Read more...
Active Adversary Playbook 2021
Issue 3 2021 , Cyber Security
Sophos has released its Active Adversary Playbook 2021, detailing cyber-attacker behaviours and the tools, techniques and procedures Sophos’ frontline threat hunters and incident responders saw in the wild.

Read more...
Passwords are 60, time for them to go
Issue 3 2021 , Access Control & Identity Management, Cyber Security
It has been 60 years since passwords were first used at MIT and if the number of breaches in the news are anything to go by, we are no more adept at managing our passwords than we were in 1961.

Read more...
Top 10 security misperceptions
Issue 3 2021 , Cyber Security, Security Services & Risk Management
The Sophos Rapid Response team has compiled a list of the most commonly held security misperceptions they’ve encountered in the last 12 months while neutralising and investigating cyberattacks in a wide range of organisations.

Read more...
Surveillance business models are changing
CCTV Handbook 2021, Technews Publishing, Eagle Eye Networks, Bosch Building Technologies, Dahua Technology South Africa, Genetec , Editor's Choice, CCTV, Surveillance & Remote Monitoring, Cyber Security, Integrated Solutions, IT infrastructure
The CCTV Handbook round table highlighted the changes that are happening in the surveillance and security world in general, from cloud to costing models and of course, cybersecurity.

Read more...
Building walls is not enough
Issue 3 2021, Vox , Cyber Security
Cybersecurity: businesses are still building walls to defend against nuclear attacks.

Read more...
VMware releases 2021 Global Security Insights Report
Issue 3 2021 , Cyber Security
VMware releases 2021 Global Security Insights Report detailing the surge in cyberattacks targeting the ‘anywhere’ workforce, highlighting opportunity for security leaders to rethink and transform cybersecurity strategies.

Read more...
Awareness training is key to e-mail security
Issue 3 2021 , Cyber Security
Awareness training should be actively deployed to complement organisations’ efforts to secure their e-mail systems, because hackers are increasingly targeting people as security technologies become more effective.

Read more...