Quantum computers existing today are limited versions of the futuristic quantum computers that we hope to achieve in the future. However, scientists have created the hardware for the “probabilistic computer” – a device to bridge the gap between the standard PCs of today and the genuine quantum computers. The study appears in the Nature journal.
This probabilistic computer can solve quantum problems using a special trick. It uses a p-bit which is described by the research team as “poor man’s qubit”. In classical computing, a bit can either take the value 1 or 0, while qubits can take both of these values at the same time as per the laws of quantum computing. Meanwhile, the p-bit can take only 1 or 0 at a time, but the switch between two states occurs very quickly. Using the fluctuations properly, researchers can tackle the problems that are considered quantum computing problems without using a real quantum computer.
In addition to this, the p-bit can operate at room temperature whereas the qubits need super-cold conditions for their operation. P-bits can be easily adapted to the existing computers. Supriyo Datta, an electrical engineer at Purdue University, in Indiana, said that there are a group of problems solved with the help of qubits that can also be solved by the p-bits. Hence getting the name “poor man’s qubit”.
The result of the research has been a modified magnetoresistive random access memory(MRAM) device for storing information in the computers of the present day. Magnetic orientations are used to represent 0s or 1s using states of resistance. Eight custom-made MRAM p-bit units were put with a controller chip to create a probabilistic computer – where units are used to take a specific value.
Scientists were able to solve the integer factorization problems, which are usually considered quantum problems. It can also be solved by classical computers however with lesser efficiency. The probabilistic computer along with p-bits represents a middle ground between two ends. Scientists feel that the fully developed p-bit computers would solve integer factorization problems with lesser energy and time than the computers of the present day.
Ahmed Zeeshan Pervaiz, Purdue University said that the circuit occupies the same area as that of a transistor but performs the function which would take several thousand transistors to perform. The calculation speed could also be increased by parallel operation of a huge number of p-bits.
For the practical use of these machines, there is a need for more refining which would not take much time. After that, these can handle certain problems until the final leap in quantum computing occurs. Connecting qubits for practical use is a tough challenge until then p-bits can be used for machine learning and optimization problems.
Journal Reference: Nature