IOS Press’ Post

"🚀 Exciting News! 🚀 Our paper titled 'DAST-Net: Dense visual attention augmented spatio-temporal network for unsupervised video anomaly detection' has been published in Neurocomputing. You can find the details of our work in the following link: (https://lnkd.in/ggKfE_yd). #ResearchPublication #Neurocomputing #AnomalyDetection #UnsupervisedLearning #VideoAnalysis"

Artificial intelligence decodes the brain's intelligence pathways - PsyPost: Artificial intelligence can predict intelligence by analyzing brain connectivity. General intelligence was most accurately predicted, ... http://dlvr.it/TGxczt

How do you make sense of activity data from thousands of neurons? Researchers from the Stringer and Pachitariu labs at Howard Hughes Medical Institute (HHMI)’s Janelia Research Campus developed a new visualization tool called Rastermap to help scientists uncover activity patterns in large-scale neural recordings.  Learn more here: https://lnkd.in/eFRxK7pe #MovingScienceForward #ScienceNews #HHMIJanelia

The Neuron Tracing and Active Learning Environment (NeuroTrALE), an #RD100 award-winning open-source software, uses artificial intelligence techniques to create high-resolution maps of the brain's network of neurons from high-dimensional biomedical data. Learn more about NeuroTrALE: https://ow.ly/lm8f50U9olL

The human brain segments event boundaries (separation of continuous natural breakpoints), providing a neural mechanism for similar boundaries found in behavioral research. Event boundaries make me think of hopscotch, in the way the ground is segmented into events to play a game but is more accurately applied to time segmentation. In the attached study, the authors found that the brain was able to segment events in time and that they were characterized by specific activity patterns about 250-450 milliseconds after the event, the event boundary. #eeg #eventboundaries #scientificcommunication See more information in this article: https://lnkd.in/gyqWdfQM Image credit to vectorstock images: https://lnkd.in/gcWdhcib

Valletta, Malta. July 17-19, 2024 The 2nd World Conference on eXplainable Artificial Intelligence The World Conference on Explainable Artificial Intelligence is an annual event that aims to bring together researchers, academics, and professionals, promoting the sharing and discussing of knowledge, new perspectives, experiences, and innovations in eXplainable Artificial Intelligence (XAI). This event is multidisciplinary and interdisciplinary, bringing together academics and scholars of different disciplines, including Computer Science, Psychology, Philosophy, and Social Science, to mention a few, and industry practitioners interested in the practical, social and ethical aspects of the explanation of the models emerging from the discipline of Artificial intelligence (AI).

DeepMind researchers discover impressive learning capabilities in long-context LLMs In their study, the DeepMind researchers investigated how many-shot ICL affects the performance of LLMs in downstream tasks.Read More https://ift.tt/kZ6zC4M https://ift.tt/OntUFf9

Fcdnet: Fuzzy Cognition-based Dynamic Fusion Network for Multimodal Sentiment Analysis https://lnkd.in/d6mstJya The Fcdnet dynamically integrates #sentiment #scores across different modalities using a fuzzy cognition fusion mechanism (FCM), significantly enhancing the accuracy of identifying divergent sentiments across modalities. Additionally, a disparity balancing module (DBM) is proposed to normalize the representations between different modality features by penalizing the similarity of sentiments with different #degrees and rewarding the #separability of sentiments with same degree. #Experimental results demonstrate that Fcdnet outperforms state-of-the-art methods on public datasets, validating the superiority and effectiveness. Marc Polo (Ph.D) Miriam Diez Bosch PhD Josep Maria Picola Meix (Ph. D.) Víctor Curto (Ph.D) Blanquerna - Universitat Ramon Llull

New IEEE Transactions on Molecular, Biological and Multi-Scale Communications article Our article Synchronized Relaying in Molecular Communication: An AI-based Approach using a Mobile Testbed Setup has been accepted for publication in IEEE Transactions on Molecular, Biological and Multi-Scale Communications. In this paper, we propose using a reinforcement learning (RL)- based synchronizer to continually adapt a decoding threshold and detect transmitted synchronization frames in a dynamic molecular communication environment. Link to article: https://lnkd.in/dijxwtUk

I finished reading a book that was recommended by a panel member/fellow at the BIED Society AI Conference last fall. With tech being exponential in development, it gets harder and harder to stay on top of policy - gaps certainly remain. "We can't yet know all the risks and benefits that will come from the ongoing project of transhumanism. But we can continuously assess both the scientific progress and the risks it entails to maximize the benefits of neurotechnology for humanity" (Farahany, 2023, p. 210). Good read, I highly recommend - some valid concerns and implications moving forward. Makes one want to protect (or at least be mindful) of the last bastion of free thought...your actual brain housing group. The book highlights many potential benefits as well.