Tiberiu Boros

Software Developer / Computer Scientist Adobe

BIOGRAPHY

Tiberiu Boros is a Ph.D. in computer science, specifically in the field of Text-to-Speech (TTS) Synthesis. He is currently working for Adobe Systems Romania and is an associate of the Research Institute for Artificial Intelligence of the Romanian Academy. Additionally, he maintains two Machine Learning open source projects (TTS-Cube and NLP-Cube) and is a contributor to the DyNet Machine Learning Framework (developed by Carnegie Mellon University and many others). His research is focused on applied Natural Language and Speech Processing.

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 Marius Manica, Cyber Incident Response at Adobe

Are you the next cyber security superstar?

If you are passionate about an information security topic or you have strong technical skills developing researches on your own, you should definitely Apply at Call for Papers. By submitting you will have the chance to showcase your work to +2000 attendees.

Other speakers joining this year

Ionel Chirita

Threat Intelligence Manager Signify

Gabriel Ciubotaru

Lead Security Researcher Garrett - Advancing Motion

Sergey Ovchinnikov

Expert Researcher JSC Advanced Monitoring

Ready for this year's presentations?

By registering you will unlock access to 60+ speakers and two full days with cyber security news & showcases from worldwide leaders.

SPEAKERS
0
COUNTRIES
0
ATTENDEES
0
HACKING
COMPETITIONS
0
COMPANIES
0

Sponsors & Partners

They help us make this conference possible.

POWERED BY

Orange Romania is the leader of the local telecom market and part of the Orange Group, one of the largest global telecommunications operators, connecting hundreds of millions of customers worldwide. With over 11 million customers and an annual turnover exceeding 1.5 billion euros, Orange Romania connects 1 in 2 Romanians and offers an extensive range of communication solutions to its customers, both individual users and companies, from basic services up to complete voice services, fixed and mobile data, TV services or smart home services, but also mobile financial services. Orange is also a leader in innovation investing yearly over 200 million euros in network infrastructure and R&D initiatives in Romania. In the past 3 years Orange has launched two 5G Labs in Bucharest and Iasi, that aim to support researchers, startups and companies to test their 5G solutions in advance. In addition, Orange is a long-term supporter of the startup ecosystem through the Orange Fab accelerator program designed to support entrepreneurs in the development of innovative products and their distribution locally and internationally. 

Orange Services was created in 2013 and is a 100% owned subsidiary of Orange Group. As a technology services company, our DNA is in IT, but our teams also work in other domains including mobile networks and a number of commercial and business functions. Orange Services is one of the largest technology hubs in the Orange Group, working internationally for both Orange corporate functions and country operations. Through a unique combination of cutting edge know-how and expertise, our teams provide a broad range of services: development and supervision of IT services in domains such as Big Data, Cloud, M2M, IoT, TV, Connected Objects; design and development of IT infrastructure and desktop solutions; testing & planning for mobile networks; implementation of supply chain solutions and also improvement of commercial & business performance including BI, CRM, Analytics, Digital learning and Customer Care. Visit us on LinkedIn.

Gold Partners

Silver Partners

 

        

Bronze Partner

 

HACKING VILLAGE PARTNERS

COMMUNITY & MEDIA PARTNERS