Deja Vu: Salesforce Customers Hacked Again, Via GainsightdarkreadingNate Nelson, Contributing Writer
In a repeat of similar attacks during the summer, threat actors affiliated with the ShinyHunters extortion group used a third-party application to steal organizations’ Salesforce data. Read More
In a potential gift to geopolitical adversaries, the encrypted messaging app uses a leaky custom protocol that allows message replays, impersonation attacks, and sensitive information exposure from chats. Read More
Task Specific Sharpness Aware O-RAN Resource Management using Multi Agent Reinforcement Learningcs.AI updates on arXiv.org arXiv:2511.15002v1 Announce Type: new
Abstract: Next-generation networks utilize the Open Radio Access Network (O-RAN) architecture to enable dynamic resource management, facilitated by the RAN Intelligent Controller (RIC). While deep reinforcement learning (DRL) models show promise in optimizing network resources, they often struggle with robustness and generalizability in dynamic environments. This paper introduces a novel resource management approach that enhances the Soft Actor Critic (SAC) algorithm with Sharpness-Aware Minimization (SAM) in a distributed Multi-Agent RL (MARL) framework. Our method introduces an adaptive and selective SAM mechanism, where regularization is explicitly driven by temporal-difference (TD)-error variance, ensuring that only agents facing high environmental complexity are regularized. This targeted strategy reduces unnecessary overhead, improves training stability, and enhances generalization without sacrificing learning efficiency. We further incorporate a dynamic $rho$ scheduling scheme to refine the exploration-exploitation trade-off across agents. Experimental results show our method significantly outperforms conventional DRL approaches, yielding up to a $22%$ improvement in resource allocation efficiency and ensuring superior QoS satisfaction across diverse O-RAN slices.
arXiv:2511.15002v1 Announce Type: new
Abstract: Next-generation networks utilize the Open Radio Access Network (O-RAN) architecture to enable dynamic resource management, facilitated by the RAN Intelligent Controller (RIC). While deep reinforcement learning (DRL) models show promise in optimizing network resources, they often struggle with robustness and generalizability in dynamic environments. This paper introduces a novel resource management approach that enhances the Soft Actor Critic (SAC) algorithm with Sharpness-Aware Minimization (SAM) in a distributed Multi-Agent RL (MARL) framework. Our method introduces an adaptive and selective SAM mechanism, where regularization is explicitly driven by temporal-difference (TD)-error variance, ensuring that only agents facing high environmental complexity are regularized. This targeted strategy reduces unnecessary overhead, improves training stability, and enhances generalization without sacrificing learning efficiency. We further incorporate a dynamic $rho$ scheduling scheme to refine the exploration-exploitation trade-off across agents. Experimental results show our method significantly outperforms conventional DRL approaches, yielding up to a $22%$ improvement in resource allocation efficiency and ensuring superior QoS satisfaction across diverse O-RAN slices. Read More
Google has started rolling out ads in AI mode, which is the company’s “answer engine,” not a search engine. […] Read More
In a surprise move, Google on Thursday announced that it has updated Quick Share, its peer-to-peer file transfer service, to work with Apple’s equipment AirDrop, allowing users to more easily share files and photos between Android and iPhone devices. The cross-platform sharing feature is currently limited to the Pixel 10 lineup and works with iPhone, […]
American cybersecurity firm CrowdStrike has confirmed that an insider shared screenshots taken on internal systems with hackers after they were leaked on Telegram by the Scattered Lapsus$ Hunters threat actors. […] Read More
Ever wonder how some IT teams keep corporate data safe without slowing down employees? Of course you have. Mobile devices are essential for modern work—but with mobility comes risk. IT admins, like you, juggle protecting sensitive data while keeping teams productive. That’s why more enterprises are turning to Samsung for mobile security. Hey—you’re busy, so […]
A China-nexus threat actor known as APT24 has been observed using a previously undocumented malware dubbed BADAUDIO to establish persistent remote access to compromised networks as part of a nearly three-year campaign. “While earlier operations relied on broad strategic web compromises to compromise legitimate websites, APT24 has recently pivoted to using more sophisticated vectors targeting Read […]
Grafana Labs is warning of a maximum severity vulnerability (CVE-2025-41115) in its Enterprise product that can be exploited to treat new users as administrators or for privilege escalation. […] Read More
From time to time, it can be instructive to look at generic phishing messages that are delivered to one’s inbox or that are caught by basic spam filters. Although one usually doesn’t find much of interest, sometimes these little excursions into what should be a run-of-the-mill collection of basic, commonly used phishing techniques can lead […]
