A Sneaky Hacking Tool Targeting AI Infrastructure Is Lurking in Victims' Blind Spots A new type of malware can worm deep into AI coding systems to steal data and logins--and can flip a "death switch" to destroy files and keep out real users. As AI tools proliferate and become deeply ingrained in software development around the world, new research from the cybersecurity firm Crowdstrike shows how attackers are actively targeting the AI toolchain to steal access credentials, gain deeper access to a target environment, exfiltrate sensitive data, and even destroy target files and systems--all while finding new ways to cover their tracks. Researchers discovered a worm in the wild while investigating AI software supply chain attacks. Adam Meyers, CrowdStrike's senior vice president of counter adversary work, says that the company has not yet attributed the activity to a specific actor, but that it fits into larger evolutions in how attackers like TeamPCP (which Crowdstrike tracks as "Altered Spider") and North Korean groups are targeting the AI software supply chain. "This is one of the campaigns that we've seen showing that this is an emerging attack class," Meyers tells WIRED.

That said, there often exist environments that resemble in structure (dynamics) yet provide more accessible rollouts (eg, unlimited in simulators).

Unsupervised reinforcement learning aims to acquire skills without prior goal representations, where an agent automatically explores an open-ended environment to represent goals and learn the goal-conditioned policy. However, this procedure is often time-consuming, limiting the rollout in some potentially expensive target environments. The intuitive approach of training in another interaction-rich environment disrupts the reproducibility of trained skills in the target environment due to the dynamics shifts and thus inhibits direct transferring. Assuming free access to a source environment, we propose an unsupervised domain adaptation method to identify and acquire skills across dynamics. Particularly, we introduce a KL regularized objective to encourage emergence of skills, rewarding the agent for both discovering skills and aligning its behaviors respecting dynamics shifts. This suggests that both dynamics (source and target) shape the reward to facilitate the learning of adaptive skills. We also conduct empirical experiments to demonstrate that our method can effectively learn skills that can be smoothly deployed in target.

Acoustic matching aims to re-synthesize an audio clip to sound as if it were recorded in a target acoustic environment. Existing methods assume access to paired training data, where the audio is observed in both source and target environments, but this limits the diversity of training data or requires the use of simulated data or heuristics to create paired samples. We propose a self-supervised approach to visual acoustic matching where training samples include only the target scene image and audio---without acoustically mismatched source audio for reference. Our approach jointly learns to disentangle room acoustics and re-synthesize audio into the target environment, via a conditional GAN framework and a novel metric that quantifies the level of residual acoustic information in the de-biased audio. Training with either in-the-wild web data or simulated data, we demonstrate it outperforms the state-of-the-art on multiple challenging datasets and a wide variety of real-world audio and environments.

We examine the problem of transferring a policy learned in a source environment to a target environment with different dynamics, particularly in the case where it is critical to reduce the amount of interaction with the target environment during learning. This problem is particularly important in sim-to-real transfer because simulators inevitably model real-world dynamics imperfectly. In this paper, we show that one existing solution to this transfer problem-- grounded action transformation --is closely related to the problem of imitation from observation (IfO): learning behaviors that mimic the observations of behavior demonstrations. After establishing this relationship, we hypothesize that recent state-of-the-art approaches from the IfO literature can be effectively repurposed for grounded transfer learning. To validate our hypothesis we derive a new algorithm -- generative adversarial reinforced action transformation (GARAT) -- based on adversarial imitation from observation techniques. We run experiments in several domains with mismatched dynamics, and find that agents trained with GARAT achieve higher returns in the target environment compared to existing black-box transfer methods.