New Digital Design Spurs Scalable Probabilistic Computing

Artificial intelligence and machine learning could become dramatically more efficient, thanks to a new type of computer component developed by researchers at the University of California, Santa Barbara and Tohoku University, in collaboration with the Taiwan Semiconductor Manufacturing Company (TSMC). The technology is based on "probabilistic bits," or "p-bits", which are hardware elements that naturally fluctuate between 0 and 1. Unlike conventional digital bits, which are fixed in value, p-bits can efficiently explore many possibilities. This makes them well-suited for solving problems such as optimization and inference, tasks that are difficult for traditional computers.

Until now, most p-bit designs have required analog electronic components to control how often the output is 0 or 1. These analog parts, called digital-to-analog converters (DACs), are bulky, power-hungry, and expensive, hindering scalability. The new work offers a breakthrough: a fully digital p-bit design that removes DACs entirely.

Circuit diagrams of a conventional probabilistic bit (p-bit) (a) and the proposed DAC-free p-bit. "X" denotes a device whose state fluctuates probabilistically; in this work, it is implemented using a stochastic magnetic tunnel junction whose magnetization direction thermally fluctuates. In the conventional p-bit (a), an analog circuit, namely a digital-to-analog converter (DAC), is required to convert digital input signals into analog ones. In contrast, the proposed p-bit (b) can be constructed solely with digital circuits, such as delay circuits, which facilitate large-scale probabilistic computing hardware. ©Shunsuke Fukami, Kerem Camsari et al.

"The reliance on analog signals was holding back progress," says Shunsuke Fukami (Tohoku University). "So, we discovered a digital method to adjust the behavior of p-bits without needing the typically used big, clunky analog circuits."

Their approach uses small electronic devices called magnetic tunnel junctions (MTJs), which naturally switch between two states in a random manner. By feeding this 50/50 random bitstream into a simple digital circuit that gradually combines signals with controlled timing, the team can smoothly tune how likely the output is to be 0 or 1. Importantly, the same digital circuitry can also compensate for natural device-to-device variations in these stochastic elements, making the approach robust to manufacturing non-uniformities.

Input-output characteristics of probabilistic bits (p-bits) constructed using the proposed circuit with stochastic magnetic tunnel junctions that exhibit different characteristic fluctuation time constants (relaxation times). (a) Measurement results before compensating for device-to-device variability, and (b) results after compensation. Uniform input-output characteristics are achieved through the compensation. ©Shunsuke Fukami, Kerem Camsari et al.

This approach enables two capabilities that have long been obstacles in hardware-based probabilistic computing. First, the system updates its internal state in a self-organizing manner, meaning different elements naturally avoid interfering with each other. This allows many p-bits to work in parallel without a central controller. Second, the design allows a form of "on-chip annealing", a method for gradually narrowing down solutions, by changing basic timing settings rather than completely rewriting stored parameters.

Results of solving a combinatorial optimization problem via annealing using the proposed probabilistic-bit (p-bit) circuit. A three-dimensional spin-glass problem with 8×8×8 = 512 spins was used as an example. By gradually varying the delay time within the p-bit circuit, the effective temperature of the probabilistic bits can be controlled. Gradually cooling the system leads to convergence to the ground state, i.e., successful computation of the optimal combination. ©Shunsuke Fukami, Kerem Camsari et al.

This remarkable new design developed by the research team requires far less area and power than alternatives, while also being compatible with modern semiconductor manufacturing. The researchers expect that this advance will make probabilistic computing practical in applications ranging from artificial intelligence to logistics, scientific discovery, and future computing systems.

The findings were presented at the 71st Annual IEEE International Electron Devices Meeting (IEDM 2025) on December 10, 2025 (JST).

Publication Details:
Title: DAC-Free p-bits: Asynchronous Self-Coloring and On-Chip Annealing
Authors: Kemal Selcuk, Navid Anjum Aadit, Corentin Delacour, Jared Quintana Silva, Nihal Sanjay Singh, Haruna Kaneko, Shun Kanai, Yu-Jui Wu, Yi-Hsuan Chen, Yu-Sheng Chen, Yi Ching Ong, Kuo-Ching Huang, Harry Chuang, Hideo Ohno, Shunsuke Fukami and Kerem Y. Camsari
Journal: 71st Annual IEEE International Electron Devices Meeting (IEDM 2025)
DOI: XXXXXXXXXXXXX

/Public Release. This material from the originating organization/author(s) might be of the point-in-time nature, and edited for clarity, style and length. Mirage.News does not take institutional positions or sides, and all views, positions, and conclusions expressed herein are solely those of the author(s).View in full here.

You might also like