Before the acceleration of AI, insider risk always centered on human intent. Security teams monitored high-risk employees who were prone to downloading files before leaving the company, or negligent employees who engaged in thoughtless behavior (e.g., clicking a phishing link.) In other words, the insider risk threat model was based on people doing things they shouldn’t. AI has fundamentally changed this paradigm. The risk is no longer limited to ill-intentioned employees but also to those who wish to […]

Ver mais

arXiv:2603.21235v1 Announce Type: new Abstract: Nonrigid registration is conventionally divided into point set registration, which aligns sparse geometries, and image registration, which aligns continuous intensity fields on regular grids. However, this dichotomy creates a critical bottleneck for emerging scientific data, such as spatial transcriptomics, where high-dimensional vector-valued functions, e.g., gene expression, are defined on irregular, sparse manifolds. Consequently, researchers currently face a forced choice: either sacrifice single-cell resolution via voxelization to utilize image-based tools, or ignore the critical […]

Ver mais

arXiv:2505.18918v3 Announce Type: replace Abstract: Principal component analysis (PCA) is a key tool in the field of data dimensionality reduction. Various methods have been proposed to extend PCA to the union of subspace (UoS) setting for clustering data that comes from multiple subspaces like K-Subspaces (KSS). However, some applications involve heterogeneous data that vary in quality due to noise characteristics associated with each data sample. Heteroscedastic methods aim to deal with such mixed data quality. This paper develops […]

Ver mais

arXiv:2603.04895v1 Announce Type: new Abstract: Overparameterized ML models, including neural networks, typically induce underdetermined training objectives with multiple global minima. The implicit bias refers to the limiting global minimum that is attained by a common optimization algorithm, such as gradient descent (GD). In this paper, we characterize the implicit bias of GD for training a shallow ReLU model with the squared loss on high-dimensional random features. Prior work showed that the implicit bias does not exist in the […]

Ver mais

arXiv:2512.23087v2 Announce Type: replace-cross Abstract: Reinforcement Learning (RL) for Large Language Models (LLMs) faces a fundamental tension: the numerical divergence between high-throughput inference engines and numerically precise training engines. Although these systems share the same parameters, they produce slightly different probability distributions, creating a training-inference mismatch. We prove that the bound on the log-probability divergence arising from this mismatch scales as $(1-p)$, where $p$ is the token probability. This scaling induces a highly asymmetric effect: the bound vanishes […]

Ver mais

arXiv:2604.14355v1 Announce Type: new Abstract: Chemical reaction networks, or CRNs, are known to stably compute semilinear Boolean-valued predicates and functions, provided that all reactions are irreversible. However, this property does not hold for wet-lab implementations, as all chemical reactions are reversible, even at very slow rates. We study the computational power of CRNs under the reverse-robust computation model, where reactions are permitted to occur either in forward or in reverse up to a cutoff point, after which they […]

Ver mais

Soft Q-learning has emerged as a versatile model-free method for entropy-regularised reinforcement learning, optimising for returns augmented with a penalty on the divergence from a reference policy. Despite its success, the multi-step extensions of soft Q-learning remain relatively unexplored and limited to on-policy action sampling under the Boltzmann policy. In this brief research note, we first present a formal $n$-step formulation for soft Q-learning and then extend this framework to the fully off-policy case by introducing a novel […]

Ver mais

arXiv:2602.06088v1 Announce Type: new Abstract: We introduce a Transformer-based Reinforcement Learning framework for autonomous orbital collision avoidance that explicitly models the effects of partial observability and imperfect monitoring in space operations. The framework combines a configurable encounter simulator, a distance-dependent observation model, and a sequential state estimator to represent uncertainty in relative motion. A central contribution of this work is the use of transformer-based Partially Observable Markov Decision Process (POMDP) architecture, which leverage long-range temporal attention to interpret […]

Ver mais