January 2026

Large Language Models have come a long way — from simple chatbots that answer questions to powerful systems that can reason, plan, and interact with the real world. But even the smartest LLMs still have a limitation: they don’t act on their own. They respond when asked, and that’s where the story usually ends. This is where AI agents change the game. By combining LLMs with memory, tools, and planning abilities, AI agents can break down complex problems, make decisions, and execute […]

Ver mais

In Google Calendar’s Secret Engineer Weapon: Restraint, I praised Google for unapologetically saying “no” to the temptations of modernity. Thanks for reading Fullstack Engineer! Subscribe for free to receive new posts and support my work. Subscribe “Use a frontend framework.” No. JavaScript. “Use Typescript.” No. Javascript. “Use a trendy CSS tool.” No. CSS classnames. “Give us a native desktop app.” No. Browser. “Give us dark mode.” No. “Make it really fast.” No. Their commitment to simplicity has helped […]

Ver mais

You’ve mastered streaming text. The AI types, the user reads, and the magic of perceived performance is achieved. But what if the AI could do more than just tell? What if it could show and let you interact? Welcome to the paradigm shift. We are moving Beyond Text. In this guide, we’ll explore how to stream actual React components directly from the server to the client. Using the Vercel AI SDK and React Server Components (RSCs), we’ll transform a static chat interface into […]

Ver mais

arXiv:2510.22848v2 Announce Type: replace-cross Abstract: Self-induced stochastic resonance (SISR) is the emergence of coherent oscillations in slow-fast excitable systems driven solely by noise, without external periodic forcing or proximity to a bifurcation. This work presents a physics-informed machine learning framework for modeling and predicting SISR in the stochastic FitzHugh-Nagumo neuron. We embed the governing stochastic differential equations and SISR-asymptotic timescale-matching constraints directly into a Physics-Informed Neural Network (PINN) based on a Noise-Augmented State Predictor architecture. The composite loss […]

Ver mais

arXiv:2510.06165v2 Announce Type: replace-cross Abstract: Feature attributions are post-training analysis methods that assess how various input features of a machine learning model contribute to an output prediction. Their interpretation is straightforward when features act independently, but it becomes less clear when the predictive model involves interactions, such as multiplicative relationships or joint feature contributions. In this work, we propose a general theory of higher-order feature attribution, which we develop on the foundation of Integrated Gradients (IG). This work […]

Ver mais

arXiv:2509.18766v2 Announce Type: replace-cross Abstract: Diagonal linear networks are neural networks with linear activation and diagonal weight matrices. Their theoretical interest is that their implicit regularization can be rigorously analyzed: from a small initialization, the training of diagonal linear networks converges to the linear predictor with minimal 1-norm among minimizers of the training loss. In this paper, we deepen this analysis showing that the full training trajectory of diagonal linear networks is closely related to the lasso regularization […]

Ver mais

arXiv:2509.00923v2 Announce Type: replace-cross Abstract: Monte Carlo Counterfactual Regret Minimization (MCCFR) has emerged as a cornerstone algorithm for solving extensive-form games, but its integration with deep neural networks introduces scale-dependent challenges that manifest differently across game complexities. This paper presents a comprehensive analysis of how neural MCCFR component effectiveness varies with game scale and proposes an adaptive framework for selective component deployment. We identify that theoretical risks such as nonstationary target distribution shifts, action support collapse, variance explosion, […]

Ver mais

arXiv:2506.11743v2 Announce Type: replace-cross Abstract: Graph coarsening aims to diminish the size of a graph to lighten its memory footprint, and has numerous applications in graph signal processing and machine learning. It is usually defined using a reduction matrix and a lifting matrix, which, respectively, allows to project a graph signal from the original graph to the coarsened one and back. This results in a loss of information measured by the so-called Restricted Spectral Approximation (RSA). Most coarsening […]

Ver mais

arXiv:2504.06250v2 Announce Type: replace-cross Abstract: We study the geometric properties of random neural networks by investigating the boundary volumes of their excursion sets for different activation functions, as the depth increases. More specifically, we show that, for activations which are not very regular (e.g., the Heaviside step function), the boundary volumes exhibit fractal behavior, with their Hausdorff dimension monotonically increasing with the depth. On the other hand, for activations which are more regular (e.g., ReLU, logistic and $tanh$), […]

Ver mais