Technology Category
Machine Learning: Hardware-aware MLConstraint Satisfaction and Optimization: Distributed CSP/OptimizationSearch and Optimization: Combinatorial Optimization
Multi-Phase Coupled CMOS Ring Oscillator based Potts Machine
🤖 AI Summary
This work addresses NP-hard multivalued combinatorial optimization problems (COPs), for which conventional Ising machines—restricted to binary spins—are inadequate. We propose a hardware Potts machine architecture based on CMOS ring oscillators, enabling compact, high-dimensional spin representation. Its core innovation is the first application of N-th-order subharmonic injection locking (N-SHIL) to stabilize multiple distinct phases within a single oscillator, thereby realizing efficient mapping of Potts spins. Coupled with a scalable multiphase coupling circuit, the architecture forms a fully hardware-implemented, extensible Potts model solver. Experiments on DIMACS graph coloring and large-scale random instances (up to 2,000 nodes) demonstrate solution accuracies of 89–92%, with no degradation in accuracy as problem size increases, constant solution time, and linear power scaling. The design significantly improves energy efficiency and scalability for multivalued COPs.
Technology Category
Application Category
📝 Abstract
This paper presents a coupled ring oscillator based Potts ma chine to solve NP-hard combinatorial optimization problems (COPs). Potts model is a generalization of the Ising model, cap turing multivalued spins in contrast to the binary-valued spins allowed in the Ising model. Similar to recent literature on Ising machines, the proposed architecture of Potts machines imple ments the Potts model with interacting spins represented by cou pled ring oscillators. Unlike Ising machines which are limited to two spin values, Potts machines model COPs that require a larger number of spin values. A major novelty of the proposed Potts machine is the utilization of the N-SHIL (Sub-Harmonic Injection Locking) mechanism, where multiple stable phases are obtained from a single (i.e. ring) oscillator. In evaluation, 3 coloring problems from the DIMACS SATBLIB benchmark and two randomly generated larger problems are mapped to the pro posed architecture. The proposed architecture is demonstrated to solve problems of varying size with 89% to 92% accuracy averaged over multiple iterations. The simulation results show that there is no degradation in accuracy, no significant increase in solution time, and only a linear increase in power dissipation with increasing problem sizes up to 2000 nodes.
Problem
Research questions and friction points this paper is trying to address.
Solves NP-hard combinatorial optimization problems using Potts machines
Generalizes Ising model with multivalued spins via coupled oscillators
Achieves 89-92% accuracy in solving large-scale graph coloring problems
Innovation
Methods, ideas, or system contributions that make the work stand out.
Coupled ring oscillators represent Potts model spins
N-SHIL enables multiple stable phases per oscillator
Solves large problems with linear power increase
🔎 Similar Papers
No similar papers found.
