CUbIQ Technologies’ Post

As we “CLAWS” in on our next Tech Tuesday feature - a question for the audience: what if it were possible to combine classic engineering constructs with quantum capabilities to revolutionize how weapon science is currently conducted? What if we told you that’s happening right now thanks to the Advancement of Science and Technology Priorities, (ARAP), CLAWS team! Their technology in question? The CLassical-quantum hybrid constructs to Advance Weapon Systems, or (CLAWS)! Team Lead Dr. Jeffery Allen and crew are in essence enabling future integrated chip-scale solutions combining classical and quantum physics for practical systems with unprecedented functionality for target engagement and execution. But this capability doesn’t stop here, impacts of this will create more versatile, reconfigurable pieces of technology that can adapt to mission needs real time, all thanks to wave matter interactions, effectively taking the “periodic table into the 3rd dimension.” #AFResearchLab | #TechTuesday | #AFRLTech | #Quantum

Fermilab's communications team made a wonderful video tour of our quantum network lab. For those interested in the science of quantum communication, this video is for you! #Quantum #Entanglement #photonics

In our latest episode, discover how Terra Quantum AG is revolutionizing complex problem-solving, enhancing security through quantum key distribution, and pushing the boundaries of hybrid quantum machine learning. Join Phil Merchant, Principal at Marks & Clerk, and Vishal Shete, Managing Director of Terra Quantum UK, as they explore cutting-edge advancements in quantum technologies and their profound impact on industries ranging from financial services to life sciences and energy. Vishal shares his journey, insights, and the exciting future of quantum applications, offering a glimpse into how these innovations will reshape our world. Don't miss this fascinating conversation that connects theory with real-world solutions in the rapidly evolving quantum landscape! Watch the full episode here: https://lnkd.in/dRU-m6K5 For more exclusive interviews, cutting-edge articles, and the latest updates, check out deadcatlivecat.com. #deadcatlivecat #quantumtechnology #quantumtech #quantummechanics #quantumphysics #terraquantum #quantumcomputing #quantumnews

This demonstration showcases a ROS1 implementation of the Zero-Shot model from the VisionFactory https://lnkd.in/gFC3uMbS , running on the city sequence dataset from FusionPortableV2 https://lnkd.in/gH2hCKQF. https://lnkd.in/gzHwgZz2 The dataset features extensive and intricate real-world scenarios, highlighting the challenges associated with cross-modality learning. Thanks for the years of devotion from Weihe Xiang, Xiangcheng Hu, Jingwen Yu , Jianhao Jiao, Jin Wu and more students of our group who keep improving the datasets.

There is a great similarity between a setup to entangle two qubits (for example: 2 spin photon interfaces + optical fibers +1 beam splitter+ 2 detectors) and a Stochastic Finite State Machine (SFSM). This SFSM can be thought of as a 'machine' which generates entanglement with a certain success rate (hence stochastic) and goes through certain discrete states (hence FSM). Now the task to create a cluster state, which are entangled network of qubits, is similar to operating an ensemble of SFSMs and creating a network of these machines. The tricky part is that each SFSM has inhomogeneous properties (spin, cavity, detunings, etc). So how do we make sure that all these SFSMs work in 'unison' to give us a largest cluster state? That's where we propose an intelligent agent (Reinforcement Learning agent) which tries to take this dynamic decision of allocating quantum resources for these SFSMs. We introduce a framework utilizing a transformer model to implement this. If it seems interesting check out our recent paper. I would like to thank Prof. Dirk Englund, Prof. Kaiming He, and Dr. Linsen Li for the opportunity and guidance.

We're exploring Quantum Machine Learning (QML) to improve the speed and accuracy of predicting molecular spectral properties. Current methods like density functional theory (DFT) are reliable but have limitations, especially for complex molecules. https://lnkd.in/eVFTRUuT #quantumcomputing #quantummachinelearning

Now out in arXiv from MIT-CQE member Dirk Englund and co-authors “Dynamic Inhomogeneous Quantum Resource Scheduling with Reinforcement Learning” by Linsen Li Pratyush Anand, Kaiming He and Dirk Englund Read more: https://lnkd.in/ehaW9tKq #quantum #quantumcomputing #quantumphysics #quantumtechnology #quantumtechnologies #quantumtech #quantumcomputers #quantumcomputer #superconducting  

How DeepSouth is Building Artificial Brains André explores balanced excitation-inhibition networks in initial testing, mirroring human cortex activities, and reveals potential for large-scale simulations. #deepsouth #airesearch #neuralnetworks https://buff.ly/3QtzP4S

Who doesn't like a Shortcut? With Spatial Shortcuts in Graph Neural Controlled Differential Equations one can increase forecasting performance by leveraging prior spatial graph structure. -We introduce an advection simulation on graph edges that allows for controlled experimental setups. -We incorporate causal graph topology information via Spatial Shortcuts into a Neural Controlled Differential Equation. -Our informed NCDE with Spatial Shortcuts requires fewer parameters to reach a lower MAE with less Parameters on the simulated data. Our Paper "Spatial Shortcuts in Graph Neural Controlled Differential Equations" was accepted at the NeurIPS2024 Workshop on Data-driven and Differentiable Simulations, Surrogates, and Solvers (D3S3). Come by at our Poster in Vancouver on 15th December 2024 from 11:00AM to 12:00PM! Check out our Paper: https://lnkd.in/eEQ66ZpJ Many Thanks to all the coauthors: Gabriel Nobis Jackie Ma Wojciech Samek! #Research #Innovation #MachineLearning #NeurIPS #NeurIPS2024 #NeuralDifferentialEquations

Interested to find out more about how we are using #quantumcomputing from D-Wave to search a molecular space of a nonillion molecules? Join our CTO Bill Shipman next week at #Qubits2024 to learn about our use of binary #quadratic models to solve complex #optimization problems.