QCi's unconventional hardware
Welcome to our learning module on unconventional computing. Computers as we know them today have followed the same fundamental principles for over 75 years. Despite their success, there are still many problems that are intractable even today. These problems may require a more unconventional approach. At QCi, we are researching, developing, and commercializing several types of unconventional hardware. This module will explain why we took this approach, and provide details on a few of our products.
Module Contents
Module Overview
1. Unconventional Computing
First, we will discuss Unconventional Computing broadly. Conventional computing relies on a central processing unit (CPU) to perform logical operations in a sequential manner using binary code. Unconventional computing, on the other hand, explores radically different methods, such as quantum systems, to address tasks where traditional computers struggle, like complex optimization and machine learning. Examples include probabilistic computing using quantum mechanics and analog computing, which leverage continuous states for unique computational advantages.
2. The Quantum Zeno Effect
The Quantum Zeno Effect allows a system to interact by frequently measuring it, preventing it from changing states. This effect is based on the principle that observing a quantum system affects its behavior, effectively "freezing" it in its current state through rapid measurements. This phenomenon can be harnessed to facilitate interactions between light particles, which is crucial for quantum computing.
3. The Zeno Blockade
The Zeno Blockade leverages the quantum Zeno effect, where frequent observation prevents a quantum system from changing states, to make two photons interact by effectively "bouncing" one off the other. This interaction is facilitated using an optical cavity and sum-frequency generation, creating conditions where nonlinear effects either allow or block light particles based on specific resonance and interaction parameters.
4. Nonlinearity vs Linearity
Nonlinear systems are those in which outputs are not directly proportional to inputs, unlike linear systems where superposition holds. In fields like nonlinear optics, the behavior of light depends on intensity, leading to effects that are crucial for quantum optics. While quantum mechanics is fundamentally linear, nonlinear models are often used for practical calculations, especially when dealing with complex systems such as those involving individual particles.
4. Entropy Quantum Computing Overview
Next, we explain Entropy Quantum Computing. Generally, in quantum information processing, minimizing loss and noise is crucial but challenging due to large overheads associated with error correction. Entropy Quantum Computing (EQC) takes a different approach by harnessing loss and noise to build quantum machines with potentially superior capacity and speed.
5. Reservoir Photonic Computing Overview
Lastly, we explore QCi's approach to Reservoir Computing, our hybrid photonic-electronic architectures for AI systems using non-linear optics and optical pulses. This structure, ideal for physical implementation, simplifies training and reduces power consumption compared to traditional neural networks. Photonic reservoir computing, leveraging light for scalability and efficiency, can potentially evolve into quantum systems, offering advantages like bypassing barren plateaus in quantum machine learning.
Conclusion
Throughout this module, we examine the theoretical underpinnings of unconventional computing but discuss a few specific implementations. From entropy quantum computing to reservoir computing, we showcase how these unconventional approaches are poised to tackle some of the most pressing challenges in computation. Next, consider diving into the Introduction to Dirac-3 module, or the Introduction to Reservoir Computing module.