Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Security Management

11/27/2019
09:30 AM
Larry Loeb
Larry Loeb
Larry Loeb
50%
50%

False Training Information Can Dupe Machine Learning Models

Researchers from Boston University have shown how really small amounts of disinformation can taint the learning process used by many AI programs.

Researchers from Boston University have recently shown how really small amounts of disinformation can taint the learning process that is used by many "AI" programs.

Panagiota Kiourti, Kacper Wardega, Susmit Jha and Wenchao Li authored the paper that has come out of this effort, "TrojDRL: Trojan Attacks on Deep Reinforcement Learning Agents." The paper examines machine learning (ML) systems that are being trained with "reinforcement learning" and came up with a way to fool them so that a Trojan could be slipped into the result of the training.

Neural nets used in ML have long been known to be sensitive to the effects of any low-quality data used in training them. These so-called "adversarial examples" are slightly perturbed inputs that can cause a neural network for a classification task to classify them as a completely different category compared to the original input.

Disturbingly, these perturbed inputs can appear identical to the original from a human perspective.

Sometimes, ML machines will be trained on third-party data sets. Should an attacker gain access to such a model data set and weaponize it with a backdoor to Trojan, the effects could be immense.

The researchers set out to deliberately introduce malicious adversarial examples that would affect the ML's performance in making classifications. For their research, they used a popular and publicly available reinforcement-learning algorithm from DeepMind, called Asynchronous Advantage Actor-Critic, or A3C.

The attack methods were tested on several Atari games that were set up to function in an environment created for reinforcement-learning research. They were Breakout, Pong, Qbert, Space Invaders, Seaquest and Crazy Climber. The games were used since the researchers could measure the effects of the decision/classification performed by the ML used by them.

The attacks are performed on a machine with an Intel i7-6850K CPU and 4×NvidiaGeForce GTX 1080 Ti GPUs that typically completes one training process every 2.4 hours.

Once they tried to defend against attacks they had recognized, things got head-scratching for them. They found that, "Untargeted attacks are difficult to defend against because untargeted attack triggers induce a distribution over outputs […] an effect that breaks the assumptions of Neural Cleanse. There is no demonstrated defense for partial Trojans, where the trigger only corrupts a subset of the output labels."

If an attack is involved with a system having wide dynamic range in its training, they say a defense "will require entirely new defense techniques as all known defenses rest on the basis of discrete outputs. Furthermore, we claim that previous works promising defenses under Threat Model 2 are not effective on Trojaned DRL agents as large training sets and small amount of poisoned inputs inhibit the proper function of such techniques."

So, they can get ML systems to make major classification errors with these adversarial examples, but they are not sure how to defend against them. It makes sense for them to conclude that, "Our work suggests caution in deploying reinforcement learning in high-security safety-critical applications where the training process is not restricted to a controlled and secure environment."

— Larry Loeb has written for many of the last century's major "dead tree" computer magazines, having been, among other things, a consulting editor for BYTE magazine and senior editor for the launch of WebWeek.

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
COVID-19: Latest Security News & Commentary
Dark Reading Staff 7/13/2020
Omdia Research Launches Page on Dark Reading
Tim Wilson, Editor in Chief, Dark Reading 7/9/2020
Russian Cyber Gang 'Cosmic Lynx' Focuses on Email Fraud
Kelly Sheridan, Staff Editor, Dark Reading,  7/7/2020
Register for Dark Reading Newsletters
White Papers
Video
Cartoon
Current Issue
Special Report: Computing's New Normal, a Dark Reading Perspective
This special report examines how IT security organizations have adapted to the "new normal" of computing and what the long-term effects will be. Read it and get a unique set of perspectives on issues ranging from new threats & vulnerabilities as a result of remote working to how enterprise security strategy will be affected long term.
Flash Poll
The Threat from the Internetand What Your Organization Can Do About It
The Threat from the Internetand What Your Organization Can Do About It
This report describes some of the latest attacks and threats emanating from the Internet, as well as advice and tips on how your organization can mitigate those threats before they affect your business. Download it today!
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2020-14300
PUBLISHED: 2020-07-13
The docker packages version docker-1.13.1-108.git4ef4b30.el7 as released for Red Hat Enterprise Linux 7 Extras via RHBA-2020:0053 (https://access.redhat.com/errata/RHBA-2020:0053) included an incorrect version of runc that was missing multiple bug and security fixes. One of the fixes regressed in th...
CVE-2020-14298
PUBLISHED: 2020-07-13
The version of docker as released for Red Hat Enterprise Linux 7 Extras via RHBA-2020:0053 advisory included an incorrect version of runc missing the fix for CVE-2019-5736, which was previously fixed via RHSA-2019:0304. This issue could allow a malicious or compromised container to compromise the co...
CVE-2020-15050
PUBLISHED: 2020-07-13
An issue was discovered in the Video Extension in Suprema BioStar 2 before 2.8.2. Remote attackers can read arbitrary files from the server via Directory Traversal.
CVE-2020-10987
PUBLISHED: 2020-07-13
The goform/setUsbUnload endpoint of Tenda AC15 AC1900 version 15.03.05.19 allows remote attackers to execute arbitrary system commands via the deviceName POST parameter.
CVE-2020-10988
PUBLISHED: 2020-07-13
A hard-coded telnet credential in the tenda_login binary of Tenda AC15 AC1900 version 15.03.05.19 allows unauthenticated remote attackers to start a telnetd service on the device.