digitado

The reinforcement learning, optimization, and control communities have lost one of their greatest pioneers. Dimitri Bertsekas passed away, leaving behind a remarkable legacy that shaped generations of researchers, engineers, and practitioners. Professor Bertsekas authored some of the most influential books in dynamic programming, optimal control, optimization, and reinforcement learning, including Dynamic Programming and Optimal Control, Neuro-Dynamic Programming, and the recently updated A Course in Reinforcement Learning. His work helped establish many of the theoretical foundations that continue to […]

Ver mais

Continual learning in Large Language Models (LLMs) is hindered by the plasticity-stability dilemma, where acquiring new capabilities often leads to catastrophic forgetting of previous knowledge. Existing methods typically treat parameters uniformly, failing to distinguish between specific task knowledge and shared capabilities. We introduce Mixture of Sparse Experts for Task Agnostic Continual Learning (SETA), a framework that resolves the plasticity-stability conflict through adaptive sparse subspace decomposition into task-specific expert modules. Unlike standard updates, where tasks compete for the same […]

Ver mais

Understanding how training data shape neural network predictions is a central problem in modern learning theory. In 2020, Pedro Domingos proposed an interpolation formula valid for every model learned by deterministic gradient descent. It expresses the model’s prediction as an integral, along the optimization path, of a data-dependent kernel that aligns the model’s gradients at the test and training data. Such a first-order characterization remains valid for models trained with batch-based stochastic optimization. In this paper, we develop […]

Ver mais

Personalized virtual heart simulations face challenges in model personalization and computational cost. While neural surrogates offer state-of-the-art solutions, they typically address either efficient personalization or training generalizable models. Recent work reframes this by learning the process of personalizing a surrogate using limited subject-specific context data, through few-shot generative modeling with set-conditioned surrogates and meta-learned amortized inference. These methods, however, assume a static and diverse training distribution with known task identifiers. When new data becomes available, they require costly […]

Ver mais

Many important outcomes unfold as dynamic cascades, including product adoption, disease spread, financial distress, and information diffusion. A central challenge is to recover the hidden influence network behind these cascades. Existing methods typically assume a specific diffusion model, and their performance degrades substantially when that assumption is misspecified. We propose CascadeNet, a Jacobian-based machine learning framework for network recovery that does not require specifying a diffusion mechanism. The key idea is that the underlying influence structure can be […]

Ver mais

I’ve been working on a reinforcement learning project focused on trading using recurrent architectures, and I’ve open-sourced it for learning and discussion. Repo: https://github.com/TiwariLaxuu/Recurrent-RL-in-Trading- The idea is to explore how recurrent models (RNN/LSTM-style components) can be integrated into RL agents for financial decision-making, especially in sequential market environments. Feel free to check it out, give feedback, or suggest improvements. If you find it useful, a star would really help support the work and motivation to keep improving it. […]

Ver mais

Real estate has always been about location, timing, and relationships. In 2026, it is also about intelligence. Behind every fast property match, Softprodigy company optimized price recommendation, predictive maintenance alert, or instant lead response, there is something more powerful at work than traditional automation. There are conversational ai in real estate systems making decisions, coordinating workflows, and continuously improving outcomes. Agentic AI is no more just experimental in real estate. It is actively reshaping how firms generate revenue, […]

Ver mais

Representation learning is central to modern machine learning, enabling transitions from handcrafted features to learned embeddings, latent spaces, foundation models, world models, and digital twins. Yet most research examines how representations are optimized after a representational framework has been selected, while less attention is given to when a new level of representation becomes necessary. We introduce the Bootstrap Theory of Representational Emergence (TBER), a framework describing how new representations arise when existing ones become explanatorily insufficient. In this […]

Ver mais