Results 121 to 130 of about 5,165 (285)
A Flexible and Energy‐Efficient Compute‐in‐Memory Accelerator for Kolmogorov–Arnold Networks
Advanced Intelligent Systems, EarlyView.This article presents KA‐CIM, a compute‐in‐memory accelerator for Kolmogorov–Arnold Networks (KANs). It enables flexible and efficient computation of arbitrary nonlinear functions through cross‐layer co‐optimization from algorithm to device. KA‐CIM surpasses CPU, ASIC, VMM‐CIM, and prior KAN accelerators by 1–3 orders of magnitude in energy‐delay ...
Chirag Sudarshan +6 more
wiley +1 more source
Optimizing 3D Bin Packing of Heterogeneous Objects Using Continuous Transformations in SE(3)
Advanced Intelligent Systems, EarlyView.This article presents a method for solving the three‐dimensional bin packing problem for heterogeneous objects using continuous rigid‐body transformations in SE(3). A heuristic optimization framework combines signed‐distance functions, neural network approximations, point‐cloud bin modeling, and physics simulation to ensure feasibility and stability ...
Michele Angelini, Marco Carricato
wiley +1 more source
Quantum-dot Cellular Automata (QCA): A Survey
, 2017 10 pages 11 figures, 3 ...
Mehta, Usha, Dhare, Vaishali
openaire +2 more sources
MusicSwarm: Biologically Inspired Intelligence for Music Composition
Advanced Intelligent Systems, EarlyView.Biologically inspired swarms of frozen foundation models self‐organize to compose complex music without fine‐tuning. By coordinating through stigmergic signals, decentralized agents dynamically evolve specialized roles and adapt to solve complex tasks.
Markus J. Buehler
wiley +1 more source
Local Search and the Evolution of World Models
Topics in Cognitive Science, EarlyView., 2023 Abstract An open question regarding how people develop their models of the world is how new candidates are generated for consideration out of infinitely many possibilities. We discuss the role that evolutionary mechanisms play in this process. Specifically, we argue that when it comes to developing a global world model, innovation is necessarily ...
Neil R. Bramley +3 more
wiley +1 more source
Scalable High‐Force Zipping Electrostatic Actuators
Advanced Intelligent Systems, EarlyView.An electrostatic actuator architecture is presented that combines high energy density with forces and strokes suited for wearable soft robotics. By folding patterned thin films into a scalable honeycomb of zipping units, performance becomes tunable through series–parallel design, achieving over 40 N and 28 J/kg.
Fabio Caruso +2 more
wiley +1 more source
Advanced Intelligent Systems, EarlyView.ResearchConnect is an AI‐powered platform that automates researcher profiling, interdisciplinary team formation, and early‐stage research ideation. By extracting keywords from papers and web sources, it quickly clusters researchers into coherent teams and generates collaborative ideas using large language models. Validation on NSF‐funded projects shows
Akshay Vilas Jadhav +2 more
wiley +1 more source
MTCA‐Net: Multi‐Task Cascade Analysis Network for Real‐Time Sperm Quality Analysis
Advanced Intelligent Systems, EarlyView.This article proposes MTCA‐Net, a multi‐task cascaded analysis network for real‐time sperm quality assessment in intracytoplasmic sperm injection. The framework integrates detection, tracking, and segmentation modules to jointly analyze sperm morphology and motility.
Jiajin Li +10 more
wiley +1 more source
Diatom‐Inspired 1D Immobile Robots Capable of 2D Collective Mobility
Advanced Intelligent Systems, EarlyView.This study presents a diatom‐inspired robotic system that explores group coordination through limited physical interactions. The researchers tune groups of Barbots, simple robotic agents that possess neither individual mobility nor explicit communication capabilities, to achieve complex and adaptive collaboration based on environmental light.
Tianyi Hu +4 more
wiley +1 more source
Biotechnology and Bioengineering, EarlyView.ABSTRACT The growing demand for biopharmaceutical products reflects their effectiveness in medical treatments. However, developing new biopharmaceuticals remains a major bottleneck, often taking up to a decade before market approval. Machine learning (ML) models have the potential to accelerate this process, but their success depends on access to large
Mohammad Golzarijalal +2 more
wiley +1 more source