A framework that uses both audio and image modality for event classification– tested on COVID-19 prescreening and battlefield object detection, later evaluated on the Raspberry Pi 4 single-board computer.
Tagged with
neural network
Latest Posts
In this episode, we talk about how a group of researchers were inspired by the adaptive immune system found in humans to fortify vulnerable neural networks and a joint effort between universities to create electric skin with unmatched performance.
The adaptive immune system serves as a template for defending neural nets from confusion-sowing attacks.
4 minutes read
Designed to address the OOD data false prediction by adaptively synthesizing virtual outliers that can maintain the model’s decision boundary during training.
Over the last decade, deep neural networks have emerged as the solution to several AI complex applications from speech recognition and object detection to autonomous vehicular systems.
By reconfiguring neural networks in artificial intelligence (AI) devices, a multi-institute team that included Penn State researchers facilitated AI systems to continually learn and adapt new data and tasks in ways that were not possible or practical before.
4 minutes read
In this episode, we talk about how TUM researchers are trying to model the effects of climate change on forest fires using a neural network, why robots assisting with getting dressed is more challenging than it seems, how drones are being used to evacuate elderly in nursing homes during emergencies.
2 minutes read
Theory-based residual neural network combines discrete choice models and deep neural networks, long viewed as conflicting methods.
In this episode, we talk about how DeepMind’s AlphaFold algorithm solved protein folding, NASA JPL’s partnership with the Department of Homeland Security to better protect firefighters, and a trial for autonomous shuttles in DTU that can change the transportation infrastructure of the whole country
Deep learning algorithms reveal the rules of gene regulation
Duke team disentangles neural networks to understand how they see the world
A new approach to image search uses images returned by traditional search methods as nodes in a graph neural network through which similarity signals are propagated, achieving improved ranking in cross-modal retrieval.
14 minutes read
Article 7 of our Cobot Series: Cobots have already staked their ground as a “must have” for industrial automation, but they’ve really just scratched the surface of what they can achieve.
By building the enormous network of wireless sensors that forms the IoT, we’re generating more data than ever before. More data promises better decision making – but only if we don’t drown in information first.
3 minutes read
Ultra-low-power MCU’s and highly efficient purpose-built chips are bringing voice to the edge.
