Having more than 7 years of experience in security, Marius has had the opportunity to see all of the branches of the industry – from state institutions to state-sponsored attacks, from private companies to privacy incidents, from law enforcement to security analyst to threat hunter and incident responder. Having had this level of exposure in such a short time has allowed him to learn faster and faster and keep up with the industry. Anyone that worked with him would describe Marius as a professional first and foremost. His dedication can be seen from his highly-focused work ethic and attention to details.

Project SCOUT. Deep Learning for malicious code detection

The number of client-side attack vectors has increased dramatically in the last decade. From exploiting browser vulnerabilities to miners or drive-by downloads, attackers commonly use Javascript code to achieve their goals. In the past, malicious code classification has been achieved using standard feature-engineering over static code analysis or dynamic code execution patterns.
We propose a new deep-learning inspired methodology for detecting malicious code, based on latent representations computed in an un-supervised manner. We explore three different methodologies for computing the latent representations in a deep encoder-decoder architecture: self-attention, global style tokens (GST) and “memory-based” representations.
The three strategies for computing latent representations capture different aspects of how the code is written: (a) the GST tokens capture specific attacker techniques like code that is obfuscated or encrypted or that does many string manipulations; (b) the memory-based method learns “code patterns” such as iterators, if/else statements, asserts etc. and (c) the multi-head attention method captures on-the-fly summarizations of code-segments that are hard to reconstruct (don’t follow standard patterns).
1. The self-attention model represents code as the concatenated values of all heads in a multi-head attention system;
2. The GST method computes a probability distribution (attention) over a fixed number of style tokens (embeddings) and the latent representation is obtained as the weighted sum over all the tokens;
3. Finally, the memory-based method is similar to GST, but it computes multiple probability distributions over different buckets of style-tokens.

The latent code representations are used as input for a multilayer perceptron that classifies a code segment as being malicious or not. Our initial experiments on previously unseen data show state-of-the art results in classifying both isolated code-sequences as well as entire JS files as being malicious or benign.

The same latent-representation extraction methodology can be used over multiple datasets, regardless of the programming language, to attend a wide-variety of code-related tasks or problems as: identifying vulnerable code, identifying bad practices, indexing code (finding similar code), copyright issues, etc.
This talk is co-presented with Tiberiu Boros, Computer Scientist at Adobe