DARPA使用人工智能創(chuàng)建現(xiàn)實(shí)環(huán)境并訓(xùn)練網(wǎng)絡(luò)智能體以應(yīng)對(duì)高級(jí)持續(xù)性網(wǎng)絡(luò)威脅

在保護(hù)關(guān)鍵計(jì)算資產(chǎn)方面，不斷擴(kuò)大的網(wǎng)絡(luò)攻擊面、頻繁的計(jì)算機(jī)漏洞掃描和繁重的安全程序造成了一場(chǎng)看似不平衡的戰(zhàn)斗。將這些因素與通常缺乏可操作反饋的昂貴的網(wǎng)絡(luò)安全評(píng)估相結(jié)合，有利于網(wǎng)絡(luò)攻擊者。

DARPA打算通過一個(gè)專注于技術(shù)的CASTLE項(xiàng)目來改變這種態(tài)勢(shì)，該項(xiàng)目可以通過自動(dòng)化、可重復(fù)和可測(cè)量的方法加速網(wǎng)絡(luò)安全評(píng)估。

用于安全測(cè)試和學(xué)習(xí)環(huán)境的網(wǎng)絡(luò)智能體(Cyber Agents for Security Testing and Learning Environments，CASTLE)項(xiàng)目旨在通過開發(fā)一個(gè)工具包來改進(jìn)網(wǎng)絡(luò)測(cè)試和評(píng)估，該工具包可實(shí)例化現(xiàn)實(shí)網(wǎng)絡(luò)環(huán)境并訓(xùn)練AI智能體以防御高級(jí)持續(xù)性網(wǎng)絡(luò)威脅(advanced persistent cyber threats，APT)。團(tuán)隊(duì)將使用強(qiáng)化學(xué)習(xí)方法來自動(dòng)化減少網(wǎng)絡(luò)漏洞的過程。

DARPA信息創(chuàng)新辦公室的CASTLE項(xiàng)目經(jīng)理Tejas Patel說：“攻擊者通常比防御者更了解網(wǎng)絡(luò)漏洞”“強(qiáng)化學(xué)習(xí)可以創(chuàng)建和培訓(xùn)網(wǎng)絡(luò)智能體，這些智能體比當(dāng)前解決網(wǎng)絡(luò)中APT的手動(dòng)方法更有效?！?/p>

CASTLE的另一個(gè)目標(biāo)是創(chuàng)建開源軟件，幫助網(wǎng)絡(luò)防御者預(yù)測(cè)攻擊者可能利用的漏洞。作為一項(xiàng)重要的好處，CASTLE軟件創(chuàng)建的數(shù)據(jù)集將促進(jìn)對(duì)超出程序生命周期的防御方法進(jìn)行開放、嚴(yán)格的評(píng)估。

更多信息也可以在CASTLE廣泛的機(jī)構(gòu)公告中找到。

https://www./news-events/2022-10-24

DARPA’s CASTLE to Fortify Computer Networks

DARPA accepting proposals using AI to create realistic environments and train cyber agents to counter advanced persistent cyber threats

An ever-expanding cyber-attack surface, infrequent computer vulnerability scans, and burdensome security procedures create a seemingly lopsided battle when it comes to defending critical computing assets. Couple those factors with costly cybersecurity assessments that often lack actionable feedback, and the odds may appear to favor bad actors.

DARPA intends to change that dynamic through a new program focused on technology that can accelerate cybersecurity assessments with automated, repeatable, and measurable approaches.

The Cyber Agents for Security Testing and Learning Environments (CASTLE) program seeks to improve cyber testing and evaluation by developing a toolkit that instantiates realistic network environments and trains AI agents to defend against advanced persistent cyber threats (APTs). Teams will use a class of machine learning known as reinforcement learning to automate the process of reducing vulnerabilities within a network.

“Attackers often have a better understanding of network vulnerabilities than defenders but it doesn’t have to be that way,” said Tejas Patel, CASTLE program manager in DARPA’s Information Innovation Office. “Reinforcement learning may enable the creation and training of cyber agents that are much more effective than current manual approaches for addressing APTs in networks.”

Another goal of CASTLE is to create open-source software that can help network defenders anticipate vulnerabilities an attacker may exploit. As an important benefit, datasets created by the CASTLE software will promote open, rigorous evaluation of defensive approaches that last beyond the life of the program.

More information can also be found in the CASTLE Broad Agency Announcement.