Kipu Quantum

𝗕𝗿𝗮𝗻𝗰𝗵-𝗮𝗻𝗱-𝗕𝗼𝘂𝗻𝗱 𝗗𝗶𝗴𝗶𝘁𝗶𝘇𝗲𝗱 𝗖𝗼𝘂𝗻𝘁𝗲𝗿𝗱𝗶𝗮𝗯𝗮𝘁𝗶𝗰 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗢𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻 Today appeared our ArXiv manuscript introducing a brand-new quantum algorithm: Branch-and-Bound Digitized Counterdiabatic Quantum Optimization (BBB-DCQO) 𝗕𝗕𝗕-𝗗𝗖𝗤𝗢 successfully runs on IBM's superconducting quantum processors at quantum-advantage level. The results? 𝗚𝗿𝗼𝘂𝗻𝗱𝗯𝗿𝗲𝗮𝗸𝗶𝗻𝗴! ✅ By integrating the rigor of classical 𝗯𝗿𝗮𝗻𝗰𝗵-𝗮𝗻𝗱-𝗯𝗼𝘂𝗻𝗱 techniques with the quantum precision of our advanced 𝗕𝗙-𝗗𝗖𝗤𝗢 𝗮𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺, BBB-DCQO outperformed both classical simulated annealing and quantum annealing approaches— solving even denser 100-variable HUBO problems with up to 𝟱𝟬𝘅 𝗳𝗲𝘄𝗲𝗿 𝗳𝘂𝗻𝗰𝘁𝗶𝗼𝗻 𝗲𝘃𝗮𝗹𝘂𝗮𝘁𝗶𝗼𝗻𝘀. 📉Unlike other methods that require costly mappings and approximations, BBB-DCQO operates directly on the original HUBO formulation, delivering exact solutions with fewer qubits and quantum gates, meaning a reduced computational overhead. 💪 On IBM’s Marrakesh quantum processor, we obtained, as compared to advanced simulated annealing (SA) and D-Wave's quantum annealing solution (QA): • Optimal results with 7M vs. 35M evaluations (BBB-DCQO vs. SA) • Class of instances that were solved with 5.1M vs. 250M evaluations (BBB-DCQO vs. SA) • Landslide solutions with 100% vs. 94% accuracy (BBB-DCQO vs. QA) This is 𝗻𝗼𝘁 𝗮 𝘁𝗵𝗲𝗼𝗿𝗲𝘁𝗶𝗰𝗮𝗹 𝗰𝗹𝗮𝗶𝗺—this is 𝗿𝗲𝗮𝗹 𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲 𝗼𝗻 𝗿𝗲𝗮𝗹 𝗾𝘂𝗮𝗻𝘁𝘂𝗺 𝗵𝗮𝗿𝗱𝘄𝗮𝗿𝗲 𝗼𝘂𝘁𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗶𝗻𝗴 𝘁𝗵𝗲 𝗯𝗲𝘀𝘁 𝗰𝗹𝗮𝘀𝘀𝗶𝗰𝗮𝗹 𝘀𝗼𝗹𝘃𝗲𝗿𝘀. In a world where industries must navigate increasingly complex optimization challenges, our work signals a paradigm shift: quantum isn’t just a promise—it’s a powerful tool for today’s hardest problems running on commercial quantum computers at quantum-advantage level. 🖥️ Read more in our blog article: ( link in the first comment) Enrique Solano, Narendra Hegade, Anton Simen, Sebastián V. Romero, Alejandro Gómez Cadavid

𝗔𝗻𝗻𝗼𝘂𝗻𝗰𝗶𝗻𝗴 𝗮 𝗦𝘁𝗿𝗮𝘁𝗲𝗴𝗶𝗰 𝗔𝗱𝗱𝗶𝘁𝗶𝗼𝗻 𝘁𝗼 𝗢𝘂𝗿 𝗟𝗲𝗮𝗱𝗲𝗿𝘀𝗵𝗶𝗽 𝗧𝗲𝗮𝗺: 𝗗𝗿𝗶𝘃𝗶𝗻𝗴 𝘁𝗵𝗲 𝗖𝗼𝗺𝗺𝗲𝗿𝗰𝗶𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝗼𝗳 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗖𝗼𝗺𝗽𝘂𝘁𝗶𝗻𝗴 The biggest inflection point in the quantum computing sector is here. At Kipu Quantum, our technology is already demonstrating 𝗾𝘂𝗮𝗻𝘁𝘂𝗺-𝗮𝗱𝘃𝗮𝗻𝘁𝗮𝗴𝗲 𝗹𝗲𝘃𝗲𝗹 performance in real-world use cases — 𝗼𝘂𝘁𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗶𝗻𝗴 𝗯𝗼𝘁𝗵 𝗰𝗹𝗮𝘀𝘀𝗶𝗰𝗮𝗹 𝘀𝗼𝗹𝘃𝗲𝗿𝘀 𝗮𝗻𝗱 𝗰𝗼𝗺𝗽𝗲𝘁𝗶𝗻𝗴 𝗾𝘂𝗮𝗻𝘁𝘂𝗺 𝗮𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺𝘀. As we enter the commercialization phase, we are pleased to welcome Eugene Oleynik as our 𝗛𝗲𝗮𝗱 𝗼𝗳 𝗦𝗮𝗹𝗲𝘀 — a strategic leader joining us to accelerate our commercial expansion. Eugene brings over 15 years of experience driving sales and go-to-market strategies across enterprise software, SaaS, and deep tech like Hewlett Packard Enterprise, Microsoft and Panaya. His expertise spans EMEA markets and beyond, with a focus on complex, high-value sales cycles — precisely the kind of leadership that transform pioneering technology into tangible market impact. “𝗧𝗵𝗶𝘀 𝗶𝘀 𝗻𝗼𝘁 𝗷𝘂𝘀𝘁 𝗮𝗻𝗼𝘁𝗵𝗲𝗿 𝗵𝗶𝗿𝗲 —𝗶𝘁’𝘀 𝗮 𝘀𝗶𝗴𝗻𝗮𝗹 𝗼𝗳 𝗼𝘂𝗿 𝗰𝗼𝗺𝗺𝗲𝗿𝗰𝗶𝗮𝗹 𝗶𝗻𝘁𝗲𝗻𝘁,” said Daniel Volz CEO of Kipu Quantum. “Our customers are asking for impact, scale, and clarity on how quantum fits their business. Eugene is exactly the kind of leader who understands how to navigate complex customer journeys, close strategic deals, and establish quantum as a core driver of enterprise innovation.” „Now that we have proven several quantum-advantage level industry use cases on today’s commercial processors, the time to scale is not soon — it’s now”, said Enrique Solano, Co-CEO of Kipu Quantum. "Eugene’s mission is to take our rapidly expanding portfolio of quantum applications in - 𝗶𝗻𝗱𝘂𝘀𝘁𝗿𝗶𝗮𝗹 𝗼𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻, 𝗾𝘂𝗮𝗻𝘁𝘂𝗺 𝘀𝗶𝗺𝘂𝗹𝗮𝘁𝗶𝗼𝗻 for 𝗺𝗮𝘁𝗲𝗿𝗶𝗮𝗹 𝗱𝗲𝘀𝗶𝗴𝗻 and chemistry, 𝗾𝘂𝗮𝗻𝘁𝘂𝗺 𝗺𝗮𝗰𝗵𝗶𝗻𝗲 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴 for 𝘁𝗮𝗯𝘂𝗹𝗮𝗿 𝗱𝗮𝘁𝗮 and 𝗶𝗺𝗮𝗴𝗲 𝗰𝗹𝗮𝘀𝘀𝗶𝗳𝗶𝗰𝗮𝘁𝗶𝗼𝗻 - and connect them with the enterprise customers ready to lead.” Eugene’s track record of delivering double-digit growth around transformative technologies aligns perfectly with our next phase: making quantum computing commercially viable, operationally scalable, and financially compelling. And with Eugene on board, we are more ready than ever to be the leader in co-designed quantum computing solutions.

𝗪𝗵𝘆 𝗟𝗔𝗕𝗦 𝗠𝗮𝘁𝘁𝗲𝗿𝘀 𝗮𝗻𝗱 𝗛𝗼𝘄 𝗞𝗶𝗽𝘂 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗜𝘀 𝗣𝘂𝘀𝗵𝗶𝗻𝗴 𝘁𝗵𝗲 𝗟𝗶𝗺𝗶𝘁𝘀 𝗼𝗳 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗢𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻   In industries like telecommunications, radar, finance, and pharmaceuticals, 𝗰𝗼𝗺𝗯𝗶𝗻𝗮𝘁𝗼𝗿𝗶𝗮𝗹 𝗼𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝗽𝗿𝗼𝗯𝗹𝗲𝗺𝘀 are more than just theory—they are at the core of real-world decision-making and operational success. Let us enter the 𝗟𝗔𝗕𝗦 𝗽𝗿𝗼𝗯𝗹𝗲𝗺—a notoriously difficult optimization challenge with far-reaching practical impact.   ❓ 𝗪𝗵𝘆 𝗟𝗔𝗕𝗦 𝗠𝗮𝘁𝘁𝗲𝗿𝘀: The 𝗟𝗼𝘄-𝗔𝘂𝘁𝗼𝗰𝗼𝗿𝗿𝗲𝗹𝗮𝘁𝗶𝗼𝗻 𝗕𝗶𝗻𝗮𝗿𝘆 𝗦𝗲𝗾𝘂𝗲𝗻𝗰𝗲 (LABS) problem is essential in applications such as #radar and #digitalcommunication. Minimizing autocorrelation means clearer signals, fewer false detections, and better performance overall. Yet, even with just 𝟲𝟳 𝗯𝗶𝗻𝗮𝗿𝘆 𝘃𝗮𝗿𝗶𝗮𝗯𝗹𝗲𝘀, no classical method has cracked it optimally.   💪 𝗢𝘂𝗿 𝗖𝗼𝗻𝘁𝗿𝗶𝗯𝘂𝘁𝗶𝗼𝗻 𝗮𝘁 𝗞𝗶𝗽𝘂 𝗤𝘂𝗮𝗻𝘁𝘂𝗺: As part of the Quantum Optimization Work Group, we at Kipu Quantum took on this challenge with our Bias-field Digitized Counterdiabatic Quantum Optimization (BF-DCQO) algorithm.   👇 Results that speak for themselves: ✅ Achieved a scaling factor of ~1.26^N, significantly better than 𝗖𝗣𝗟𝗘𝗫 (~1.73^N) and 𝗚𝘂𝗿𝗼𝗯𝗶 (~1.61^N) ✅ Outperformed the best classical algorithm and matched 𝗤𝗔𝗢𝗔-12 using 6x fewer entangling gates ✅ Set 𝗮 𝗻𝗲𝘄 𝗾𝘂𝗮𝗻𝘁𝘂𝗺 𝗯𝗲𝗻𝗰𝗵𝗺𝗮𝗿𝗸 by solving LABS on 20 qubits—the highest ever on IBM hardware   This was a true team effort with leading minds from IBM, NTT, E.ON Digital Technology, and top academic institutions. Our joint benchmarking work, “Quantum Optimization Benchmark Library: The Intractable Decathlon,” lays the groundwork for fair, rigorous, and collaborative progress in the quantum era.   🖥️ Read more about LABS problem in our newest blog (link in the first comment) Narendra Hegade, Alejandro Gómez Cadavid, Enrique Solano

𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗘𝘅𝗲𝗰𝘂𝘁𝗶𝘃𝗲 𝗘𝗱𝘂𝗰𝗮𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗧𝗿𝗮𝗶𝗻𝗶𝗻𝗴 𝗖𝗼𝗺𝗲𝘀 𝘁𝗼 𝗣𝗟𝗔𝗡𝗤𝗞   At the 𝗪𝗼𝗿𝗹𝗱 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗗𝗮𝘆, we are happy to announce a powerful collaboration between Kipu Quantum and HorizonX Consulting to bring the Quantum Computing Executive Education course to the PLANQK platform — accelerating quantum readiness across Europe and beyond! 🌎 As quantum computing continues to evolve, there’s an urgent need for business and government leaders to understand its transformative potential. This course will deliver the essential knowledge and strategic insight needed to lead in a quantum-enabled future. By making this course available through PLANQK, we are empowering our community with world-class education, leadership development, and real-world applications of quantum technologies.   🚀 Explore the course: https://lnkd.in/ecScvzAW --> use code: HorizonX 🌐 Learn more about PLANQK: https://lnkd.in/eYrDVUry   💥 Adding even more excitement, we are proud to announce that Steve Suarez® Suarez, CEO of HorizonX Consulting and former Global Head of Innovation HSBC, will present during the upcoming quantum computing event organized by NTT DATA Europe & Latam in Peru.   📆 On 𝗔𝗽𝗿𝗶𝗹 𝟮𝟵 Kipu and NTT DATA jointly host a major event at NTT DATA’s offices in Peru to explore the vast potential of the quantum era. The race toward the quantum momentum has already begun — and Latin America must be at the forefront.  👉 Register now for the in-person event here: https://lnkd.in/enKWvNFE This is a unique opportunity to gain insider knowledge, witness quantum innovation in action, and connect with pioneers leading the charge into the quantum era.   Daniel Volz, Enrique Solano, Michael Falkenthal, Steve Suarez®, Juan José Miranda, Narciso Lema, Diego C. Juan Manuel Felix del Aguila, Andrea de Almeida del Solar, José Solano, David Niehaus, Joanna Folberth

𝗜𝘀 𝗗𝗶𝗴𝗶𝘁𝗮𝗹-𝗔𝗻𝗮𝗹𝗼𝗴 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗖𝗼𝗺𝗽𝘂𝘁𝗶𝗻𝗴 (𝗗𝗔𝗤𝗖) 𝘁𝗵𝗲 𝗺𝗶𝘀𝘀𝗶𝗻𝗴 𝗹𝗶𝗻𝗸 𝘁𝗼 𝗽𝗿𝗮𝗰𝘁𝗶𝗰𝗮𝗹 𝗾𝘂𝗮𝗻𝘁𝘂𝗺 𝗮𝗱𝘃𝗮𝗻𝘁𝗮𝗴𝗲?   As quantum computing matures, we see impressive progress in both #digital and #analog approaches. But what if combining these two worlds is the key to unlocking solutions faster, with fewer errors, and on today’s commercial quantum hardware?   ➡️ Why has DAQC, despite its demonstrated potential, remained largely unexplored by major hardware providers?   ➡️ Could DAQC offer a path to immediate quantum advantage in fields like materials science, drug discovery, optimization, and finance?   ➡️ How much faster could we tackle complex problems if we optimized the natural analog interactions of quantum processors alongside digital gate operations?   Recent experiments from Google and QuEra Computing Inc. suggest that 𝗗𝗔𝗤𝗖 can accelerate #simulations, improve quantum #errorcorrection, and outperform purely digital approaches — yet adoption remains rare.   ⁉️ Should superconducting, neutral atom, trapped ion, photonic, and spin-qubit hardware platforms invest more aggressively into DAQC?   🚀 At Kipu Quantum, we are not waiting for the future — we are building it: We have provided preliminary evidence of significant scaling up for optimization problems using 𝗗𝗶𝗴𝗶𝘁𝗮𝗹-𝗔𝗻𝗮𝗹𝗼𝗴 𝗖𝗼𝘂𝗻𝘁𝗲𝗿𝗱𝗶𝗮𝗯𝗮𝘁𝗶𝗰 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗢𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻 (𝗗𝗔𝗖𝗤𝗢) with #trappedions and #superconductors. We are now extending this approach across other platforms: superconducting circuits, neutral atoms, photonics, and semiconductor quantum processors. Our results show that digital-analog encodings can dramatically reduce time-to-solution and boost solution quality, outperforming classical techniques like tensor network methods, Gurobi, simulated annealing, and CPLEX. 💡 If quantum advantage is possible with digital methods, how much closer could DAQC bring us — right now?   Will hybrid digital-analog methods dominate early breakthroughs? Curious to hear your perspectives. Read our newest blog article and let us open the discussion in the comments! 👇 Photo: Annette Koroll FOTOS

𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗢𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝗕𝗲𝗻𝗰𝗵𝗺𝗮𝗿𝗸 𝗟𝗶𝗯𝗿𝗮𝗿𝘆: 𝗧𝗵𝗲 𝗜𝗻𝘁𝗿𝗮𝗰𝘁𝗮𝗯𝗹𝗲 𝗗𝗲𝗰𝗮𝘁𝗵𝗹𝗼𝗻 Excited to share our new article, "𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗢𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝗕𝗲𝗻𝗰𝗵𝗺𝗮𝗿𝗸 𝗟𝗶𝗯𝗿𝗮𝗿𝘆: 𝗧𝗵𝗲 𝗜𝗻𝘁𝗿𝗮𝗰𝘁𝗮𝗯𝗹𝗲 𝗗𝗲𝗰𝗮𝘁𝗵𝗹𝗼𝗻", published as part of the IBM Quantum Optimization Working Group! We're thrilled to present the latest collaborative results, with key contributions from Kipu Quantum alongside IBM Quantum, other industry leaders like E.ON Digital Technology, NTT DATA, T-Systems International and academic researchers. The paper, now available on arXiv (arXiv:2504.03832), explores 10 combinatorial optimization problem classes that are found to be difficult for state-of-the-art classical solvers.   A standout highlight is Kipu’s work on the 𝗟𝗼𝘄 𝗔𝘂𝘁𝗼𝗰𝗼𝗿𝗿𝗲𝗹𝗮𝘁𝗶𝗼𝗻 𝗕𝗶𝗻𝗮𝗿𝘆 𝗦𝗲𝗾𝘂𝗲𝗻𝗰𝗲 (LABS) problem — one of the challenging problem for classical methods—using our latest 𝗕𝗙-𝗗𝗖𝗤𝗢 𝗮𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺. We demonstrated quantum scaling advantages over solvers like CPLEX and Gurobi for sequences up to 30. Notably, we solved LABS instances up to 20 qubits on IBM hardware, surpassing the previous record of 18 qubits achieved on Quantinuum’s quantum computer by JPMorganChase. Our single layer BF-DCQO approach also matched the performance of a 12-layer QAOA circuit while delivering a 𝟲𝘅 𝗿𝗲𝗱𝘂𝗰𝘁𝗶𝗼𝗻 in entangling gate count, with no need for classical optimization. This benchmark study underscores the growing potential of quantum computing to outperform classical algorithms on complex optimization tasks and invites the broader community to take on the challenge—explore the provided problem instances and share your solutions! Try the most advanced quantum optimizer: Iskay from Kipu Quantum https://lnkd.in/d_vT6s-6 Thorsten Koch David Esteban Bernal Neira Ying Chen Giorgio Cortiana Daniel J. Egger Raoul Heese Narendra Hegade Alejandro Gómez Cadavid Rhea Huang Toshinari Itoko Thomas Kleinert Pedro Maciel Xavier Naeimeh Mohseni Alejandro Montanez-Barrera Koji Nakano Giacomo Nannicini Dr. Corey O'Meara Justin Pauckert Manuel Proissl @Anurag Ramesh Maximilian Schicker Noriaki Shimada @Mitsuharu T. Víctor Valls @David Van Bulck Christa Zoufal, Stefan Woerner

𝗤𝘂𝗘𝗿𝗮 𝘄𝗶𝘁𝗵 𝗙𝗿𝗶𝗲𝗻𝗱𝘀 — 𝗡𝗲𝘄 𝗪𝗲𝗯𝗶𝗻𝗮𝗿 𝗖𝗼𝗺𝗶𝗻𝗴 𝗨𝗽! Our Head of Development, Jan Trautmann, will be speaking at QuEra Computing Inc.'s upcoming webinar: "QuEra's World Tour: Germany". Jan will share insights into our application- and hardware-specific quantum computing solutions for optimization problems in #logistics, #telecommunication, and explore our latest work on feature mapping and image classification. We will also discuss new opportunities emerging in the quantum ecosystem and how our collaboration with QuEra Computing Inc. is helping to drive progress. 𝗪𝗲𝗯𝗶𝗻𝗮𝗿 𝗗𝗲𝘁𝗮𝗶𝗹𝘀: 🗓️ Monday, April 14th ⏳ 11:30 AM ET | 17:30 CET Registration link in the first comment below 👇 Register now and join Jan Trautmann, Yuval Boger, Alexandra Beckstein Markus Buchberger in the webinar next week.

KIPU DEPLOYS QUANTUM ADVANTAGE ON D-WAVE WITH ANALOG QUANTUM FEATURE MAPPING APPLIED TO BANKRUPTCY PREDICTION At Kipu Quantum, we tested bankruptcy prediction on the quantum quench dynamics of the D-Wave fast annealing regime of the Advantage2 device, where there is a runtime quantum advantage in comparison with advanced classical solvers. This is a solid case for our Analog Quantum Feature Mapping (AQFM) technology running on a highly imbalanced 95-variable dataset for predicting bankruptcy with 7,000 samples. We processed multiple samples at the same time by embedding the problems as uncoupled 95-qubit spin Hamiltonians running in the range of 10-30 nanoseconds. The dataset, spanning 1999–2009 from the Taiwan Economic Journal, classifies company bankruptcy according to Taiwan Stock Exchange business regulations. https://lnkd.in/d53YWBUn The quantum hardware experiments took a few minutes of QPU time. We performed 6,000 shots with annealing times of 20 nanoseconds. Simulating the fast quantum annealing dynamics is intractable for classical computers. Enhanced results can be achieved by increasing the density of the Hamiltonian, using more qubits, and reducing parallel embeddings.   Our results, both quantum and classical, are the best compared to the publicly available ones. The model with AQFM learns better about the minority class, with a recall of 0.77 against the classical one of 0.74, increasing the number of true positives in the bankruptcy prediction. The balanced accuracy is also slightly better with our AQFM. Why do true positives matter more? Predicting that a company will go bankrupt when it won’t, a false positive, might lead to being overly cautious, e.g. denying a loan, but it usually has lower financial risk than the reverse error. In most bankruptcy datasets, bankruptcies are a small minority class, say <1%. A model could easily achieve high accuracy by always predicting "no bankruptcy", but it would miss the actual bankruptcies, defeating the purpose of the model.   This result shows a potential quantum advantage in classification tasks by using a smart encoding of industry use cases, via Kipu Quantum technologies, on experimental regimes with quantum supremacy capabilities. The talent strongly relies on turning quantum-supremacy claims into quantum-advantage industry applications. Quantum destiny is all. We will keep our weekly postings presenting the most advanced quantum-advantage solutions with our Kipu Quantum algorithms running on commercial quantum processors in digital, analog, or digital-analog encodings. IBM IonQ Quantinuum QuEra Computing Inc. Pasqal Google Amazon Web Services (AWS) Microsoft eleQtron GmbH QUDORA Technologies Quandela Quobly IQM Quantum Computers Oxford Quantum Circuits (OQC) Alice & Bob #tariffs #stockmarket #financialcrisis #QML Anton Simen Archismita Dalal, PhD Alejandro Gómez Cadavid Narendra Hegade Juan Francisco Rodríguez Hernández

ANALOG QUANTUM CONVOLUTIONAL NEURAL NETWORKS (AQCNN) FOR IMAGE CLASSIFICATION WITH D-WAVE INTO QUANTUM ADVANTAGE At Kipu Quantum, we actively explore advancements in commercial quantum hardware, unveiling quantum-advantage level industry applications. By encoding image datasets and generating quantum-convolved representations via quantum convolutions, we boost the performance of classical machine learning. In doing so, we extend previous demonstrations of quantum supremacy in quantum hardware, https://lnkd.in/ewMaEvjF, to quantum-advantage level solutions. In contrast to our previous work on Digital-Analog Quantum Convolutional Neural Networks (DAQCNN), https://lnkd.in/dJzRaUDd, using neutral atoms, we developed a new approach for analog superconducting quantum computers. We map our algorithm in the fast-annealing regime of D-Wave Advantage2 using quantum quench dynamics. Leveraging 1,200+ qubits on it, we encode and process large-scale datasets directly on quantum hardware. On benchmark datasets for breast cancer classification, we demonstrate significant improvements compared to classical solvers. Our method reduced training parameters by a factor of 1,000, improved the Area Under Curve (AUC) by 2.72%, and increased accuracy (ACC) by 1.85% when using analog quantum convolutions on D-Wave Advantage2. The reduction in the number of parameters is achieved by replacing parameterized convolutions with non-parameterized quantum convolutions. To this end, we implement parallel convolutions, each using tens of qubits, on a 1200+ qubit chip. For higher-resolution images, we estimate to use convolutions requiring hundreds of qubits, bringing us to the level of quantum advantage. BROAD APPLICABILITY The application of AQCNN is not limited to image datasets. Its flexibility allows effective applications across different industries like stock market data, enhancing predictions by analyzing complex interactions among different variables and temporal points. A huge benefit of AQCNN is its seamless integration with existing classical machine learning models. This ensures practical usability and immediate applicability, making superior quantum solutions available today, be in finance, science, or industrial settings. We encourage researchers and industry practitioners interested in exploring this innovative approach to collaborate with us, validate our findings, and further enhance these quantum-based methods. Kipu Quantum is committed to continually demonstrating quantum-advantage level across various industries. We will continue with our weekly announcements highlighting the effectiveness and broad applicability of our quantum solutions. Read our new blog with more details: https://lnkd.in/dPJ5Ycrh Quantum destiny is all Juan Francisco Rodríguez Hernández Carlos Flores Garrigos Archismita Dalal, PhD Anton Simen Alejandro Gómez Cadavid Narendra Hegade IBM QuEra Computing Inc. IonQ Pasqal Quantinuum Google Microsoft Amazon Web Services (AWS)

𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗠𝗲𝗲𝘁𝘀 𝗔𝗜 𝗮𝘁 𝗖𝗵𝗮𝘁𝗯𝗼𝘁 𝗦𝘂𝗺𝗺𝗶𝘁 𝗕𝗲𝗿𝗹𝗶𝗻! Next week, Enrique Solano, Co-CEO of Kipu Quantum, will take the stage at Chatbot Summit to unveil how quantum computing and AI are joining forces to revolutionize the digital landscape. 👇 Key Topics Kike will Explore: ✅ How quantum computing is achieving usefulness at quantum-advantage level in today's industry use cases ✅ ChatQPT: combinatorial reasoning, HUBO quantum algorithmic mapping, and the parameterized break even from classical to quantum superintelligence ✅ Agentic Quantum Computing: AI agents solving complex use cases with sequential/parallelized homogeneous/heterogeneous quantum computers ✅ What the present and future of quantum-AI convergence means for enterprise and industry 💡 If you’re passionate about innovation, meet Enrique Solano next week at Ritz-Cartlon Berlin 📅 See you in Berlin! Registration link below 👇