Deep learning provides proactive network protection
Incremental rate of high profile threats (e.g. ransomware) up to dual–growth figure (15.8%). The result is a dangerous path that is more likely to lead to ongoing loss for organizations that fall victim to a cyber attack without any benefit in terms of defensive power. Indeed, a 2021 data breach report by IBM and the Ponemon Institute disclosure that the average cost of a data breach is $4.24 million.
In addition to cost, a cyberattack can cause irreparable damage to a company’s brand, stock price, and day-to-day operations. According to Deloitte recently survey, 32% of respondents cited an outage as the biggest impact of a network outage or breach. Other consequences cited by survey companies include intellectual property theft (22%), stock price decline (19%), loss of reputation (17%), and loss of customer trust (17). %).
Given these significant risks, organizations simply cannot accept the status quo of digital asset protection. “If we want to get ahead of our enemies, the world needs to change its mindset from detection to prevention,” Caspi said. “Organizations need to change the way they do security and fight hackers.”
Deep learning can be the difference
To date, many cybersecurity experts have viewed machine learning as the most innovative approach to digital asset protection. But deep learning is ideal for changing the way we prevent cybersecurity attacks. Any machine learning tool can be understood and theoretically reverse engineered to introduce a bias or security hole that would weaken its defenses against an attack. The bad guys can also use their own machine learning algorithms to contaminate a defensive solution with the wrong data sets.
Fortunately, deep learning addresses the limitations of machine learning by avoiding the need for highly skilled and experienced data scientists to manually provide a solution dataset. Instead, a deep learning model, developed specifically for cybersecurity, can absorb and process large volumes of raw data to fully train the system. These neural networks become autonomous, after training, and do not require constant human intervention. The combination of raw data-based learning and larger data sets means that deep learning can ultimately pinpoint much more complex patterns than machine learning, at a much faster rate.
Mirel Sehic, vice president, general manager, Honeywell Building Technologies (HBT), a multinational corporation and supplier of aerospace, performance, safety and productivity materials, said: than any rejection list, based on heuristics or standard machine learning methods. technologies. “The time it takes for a deep learning-based approach to detect a particular threat is much faster than any other factor combined.”
This content is produced by Insights, the custom content arm of MIT Technology Review. It was not written by the editorial board of the MIT Technology Review.