digitado

Diagnosing faults in modern mixed-signal system-on-a-chip (SoC) devices is a complex task. Modern chips integrate digital processors, analog signal chains, RF communication blocks, and power-management circuits on the same silicon die. Many such devices include proprietary architectures or security-sensitive intellectual property. During debugging, internal access to signals or subsystems is frequently required. Unless carefully controlled, these interfaces may unintentionally expose confidential design details. The conflict between highly accessible debugging and the need to protect intellectual property has become […]

Ver mais

Electrification of commercial automobiles is transforming the transportation industry at a pace that cannot be imagined, but it has brought with it numerous engineering challenges that extend far beyond the mere fact that batteries have to be installed where internal combustion engines used to be. Electric trucks and buses operate on large battery packs that weigh between 500 kg and 4,000 kg, which is significant for a heavy-duty vehicle and demands a lot of space, thermal control, and […]

Ver mais

Data collected from the physical world is always a combination of multiple sources: an underlying signal from the physical process of interest and a signal from measurement-dependent artifacts from the sensor or instrument. This secondary signal acts as a confounding factor, limiting our ability to extract information about the physics underlying the phenomena we observe. Furthermore, it complicates the combination of observations in heterogeneous or multi-instrument settings. We propose a deep learning framework that leverages overlapping observations, a […]

Ver mais

Rare events are central to the evolution of complex many-body systems, characterized as key transitional configurations on the free energy surface (FES). Conventional methods require adequate sampling of rare event transitions to obtain the FES, which is computationally very demanding. Here we introduce the variational free energy surface (VaFES), a dataset-free framework that directly models FESs using tractable-density generative models. Rare events can then be immediately identified from the FES with their configurations generated directly via one-shot sampling […]

Ver mais

Hi, everyone. Now I decided to try a new challenge: pixel-based SAC model for Walker2d environment. My problem is that even after a lot of training, it inmediatly falls. I have tried using optuna for hyperparameter search, but got nothing out of it. I am using stable-baselines 3 library to train it. I tried training with the by-default reward and with custom reward, but it turned out almost the same outcome: no walking at all. I do not […]

Ver mais

SusHi Tech 2026 is zeroing in on four technology domains reshaping society: AI, Robotics, Resilience, and Entertainment. Expect live demos of humanoid robots, sessions on autonomous driving’s software revolution, deep dives into cyber defense and climate tech, and candid conversations about how AI is rewriting the global music and anime industries.

Ver mais