digitado

Achieving resilient and high-quality manufacturing requires reliable data-driven anomaly detection methods that are capable of addressing differences in behaviors among different individual machines which are nominally the same and are executing the same processes. To address the problem of detecting anomalies in a machine using sensory data gathered from different individual machines executing the same procedure, this paper proposes a cross-machine time-series anomaly detection framework that integrates a domain-invariant feature extractor with an unsupervised anomaly detection module. Leveraging […]

Ver mais

For years, passwords were treated like the front door of the internet. Everything depended on them. Your email. Your bank account. Your cloud storage. Your company dashboard. Your social media. Your identity. One secret string of characters stood between you and everyone who should not be there. That system was always flawed. It survived not because it was great, but because it was simple, familiar, and cheap. Passwords were easy to create, easy to deploy, and easy for […]

Ver mais

Everyone has a version of the same fear: “AI is moving faster than I can keep up. What happens when my hard-earned skills stop mattering?” That fear isn’t irrational. But it’s also misdiagnosed. AI anxiety isn’t primarily a reaction to model capability. It’s a reaction to losing control — of learning, of work, of identity. The real shift isn’t “machines got smart.” It’s that the economy is re-pricing execution. So, if you’re trying to survive this wave by […]

Ver mais

Ordinary differential equations (ODEs) provide a powerful framework for modeling dynamic systems arising in a wide range of scientific domains. However, most existing ODE methods focus on a single system, and do not adequately address the problem of learning shared patterns from multiple heterogeneous dynamic systems. In this article, we propose a novel distributionally robust learning approach for modeling heterogeneous ODE systems. Specifically, we construct a robust dynamic system by maximizing a worst-case reward over an uncertainty class […]

Ver mais

Landauer’s principle gives a lower bound on the amount of energy it takes to erase one bit of information: E ≥ log(2) kB T where kB is the Boltzmann constant and T is the ambient temperature in Kelvin. The lower bound applies no matter how the bit is physically stored. There is no theoretical lower limit on the energy required to carry out a reversible calculation. In practice the energy required to erase a bit is around a […]

Ver mais

Modern logistics networks generate rich operational data streams at every warehouse node and transportation lane — from order timestamps and routing records to shipping manifests — yet predicting delivery delays remains predominantly reactive. Existing predictive approaches typically treat this problem either as a tabular classification task, ignoring network topology, or as a time-series anomaly detection task, overlooking the spatial dependencies of the supply chain graph. To bridge this gap, we propose a hybrid deep learning framework for proactive […]

Ver mais

Preprocessing leakage arises when scaling, imputation, or other data-dependent transformations are estimated before resampling, inflating apparent performance while remaining hard to detect. We present fastml, an R package that provides a single-call interface for leakage-aware machine learning through guarded resampling, where preprocessing is re-estimated inside each resample and applied to the corresponding assessment data. The package supports grouped and time-ordered resampling, blocks high-risk configurations, audits recipes for external dependencies, and includes sandboxed execution and integrated model explanation. We […]

Ver mais

Training dynamics during grokking concentrate along a small number of dominant update directions — the spectral edge — which reliably distinguishes grokking from non-grokking regimes. We show that standard mechanistic interpretability tools (head attribution, activation probing, sparse autoencoders) fail to capture these directions: their structure is not localized in parameter or feature space. Instead, each direction induces a structured function over the input domain, revealing low-dimensional functional modes invisible to representation-level analysis. For modular addition, all leading directions […]

Ver mais