arXiv:2601.18658v1 Announce Type: cross Abstract: When developing clinical prediction models, it can be challenging to balance between global models that are valid for all patients and personalized models tailored to individuals or potentially unknown subgroups. To aid such decisions, we propose a diagnostic tool for contrasting global regression models and patient-specific (local) regression models. The core utility of this tool is to identify where and for whom a global model may be inadequate. We focus on regression models […]

Ver mais

Aakriti Kharel, a graduate in Biology and current Master’s student in Artificial Intelligence, has traversed continents and disciplines on her path from Nepal to Texas. With professional experience in hospitals and research laboratories in both countries, and an active record of community science engagement, Kharel brings cross-cultural insight to the ongoing challenge of advancing women’s roles in STEM.  Her work foregrounds the value of empathy, leadership, and practical mentorship in an environment that remains defined by both immense […]

Ver mais

arXiv:2407.02700v4 Announce Type: replace-cross Abstract: This paper presents a novel approach for the output range estimation problem in Deep Neural Networks (DNNs) by integrating a Simulated Annealing (SA) algorithm tailored to operate within constrained domains and ensure convergence towards global optima. The method effectively addresses the challenges posed by the lack of local geometric information and the high non-linearity inherent to DNNs, making it applicable to a wide variety of architectures, with a special focus on Residual Networks […]

Ver mais

Imbalanced node classification in graph neural networks (GNNs) happens when some labels are much more common than others, which causes the model to learn unfairly and perform badly on the less common classes. To solve this problem, we propose a Curriculum-Guided Feature Learning and Three-Stage Attention Network (CL3AN-GNN), a learning network that uses a three-step attention system (Engage, Enact, Embed) similar to how humans learn. The model begins by engaging with structurally simpler features, defined as (1) local […]

Ver mais

arXiv:2603.09251v1 Announce Type: new Abstract: Learning to sample from complex unnormalized distributions is a fundamental challenge in computational physics and machine learning. While score-based and variational methods have achieved success in continuous domains, extending them to discrete or mixed-variable systems remains difficult due to ill-defined gradients or high variance in estimators. We propose a unified, target-gradient-free generative sampling framework applicable across diverse state spaces. Building on the fact that detailed balance implies the time-reversibility of the equilibrium stochastic […]

Ver mais

arXiv:2603.20298v1 Announce Type: new Abstract: A code is called solid if, roughly speaking, any correctly-transmitted codeword in an arbitrarily corrupted string of codewords can still be decoded correctly and unambiguously. So-called variable-length solid codes, in which codewords may differ in length, have been studied by various authors. In this short note, we observe that a recent construction of variable-length solid codes based on binary codes may be extended to arbitrary n-ary codes. We further prove an interesting error-detection […]

Ver mais

Here’s a dynamic I’ve seen at least a dozen times: Alice: Man that article has a very inaccurate/misleading/horrifying headline. Bob: Did you know, *actually* article writers don’t write their own headlines? … But what I care about is the misleading headline, not your org chart. Another example I’ve encountered recently is (anonymizing) when a friend complained about a prosaic safety problem at a major AI company that went unfixed for multiple months. Someone else with background information “usefully” […]

Ver mais

Generally capable agents must learn from experience in ways that generalize across tasks and environments. The fundamental problems of learning, including credit assignment, overfitting, forgetting, local optima, and high-variance learning signals, persist whether the learned object lies in parameter space or context space. While these challenges are well understood in classical machine learning optimization, they remain underexplored in context space, leading current methods to be fragmented and ad hoc. We present Reflective Context Learning (RCL), a unified framework […]

Ver mais