-
LIMO: Low-Power In-Memory-Annealer and Matrix-Multiplication Primitive for Edge Computing
Authors:
Amod Holla,
Sumedh Chatterjee,
Sutanu Sen,
Anushka Mukherjee,
Fernando Garcia-Redondo,
Dwaipayan Biswas,
Francesca Iacopi,
Kaushik Roy
Abstract:
Combinatorial optimization (CO) underpins applications in science and engineering, ranging from logistics to electronic design automation. A classic example is the NP-complete Traveling Salesman Problem (TSP). Finding exact solutions for large-scale TSP instances remains computationally intractable; on von Neumann architectures, such solvers are constrained by the memory wall, incurring compute-me…
▽ More
Combinatorial optimization (CO) underpins applications in science and engineering, ranging from logistics to electronic design automation. A classic example is the NP-complete Traveling Salesman Problem (TSP). Finding exact solutions for large-scale TSP instances remains computationally intractable; on von Neumann architectures, such solvers are constrained by the memory wall, incurring compute-memory traffic that grows with instance size. Metaheuristics, such as simulated annealing implemented on compute-in-memory (CiM) architectures, offer a way to mitigate the von Neumann bottleneck. This is accomplished by performing in-memory optimization cycles to rapidly find approximate solutions for TSP instances. Yet this approach suffers from degrading solution quality as instance size increases, owing to inefficient state-space exploration. To address this, we present LIMO, a mixed-signal computational macro that implements an in-memory annealing algorithm with reduced search-space complexity. The annealing process is aided by the stochastic switching of spin-transfer-torque magnetic-tunnel-junctions (STT-MTJs) to escape local minima. For large instances, our macro co-design is complemented by a refinement-based divide-and-conquer algorithm amenable to parallel optimization in a spatial architecture. Consequently, our system comprising several LIMO macros achieves superior solution quality and faster time-to-solution on instances up to 85,900 cities compared to prior hardware annealers. The modularity of our annealing peripherals allows the LIMO macro to be reused for other applications, such as vector-matrix multiplications (VMMs). This enables our architecture to support neural network inference. As an illustration, we show image classification and face detection with software-comparable accuracy, while achieving lower latency and energy consumption than baseline CiM architectures.
△ Less
Submitted 29 December, 2025;
originally announced December 2025.
-
TAXI: Traveling Salesman Problem Accelerator with X-bar-based Ising Macros Powered by SOT-MRAMs and Hierarchical Clustering
Authors:
Sangmin Yoo,
Amod Holla,
Sourav Sanyal,
Dong Eun Kim,
Francesca Iacopi,
Dwaipayan Biswas,
James Myers,
Kaushik Roy
Abstract:
Ising solvers with hierarchical clustering have shown promise for large-scale Traveling Salesman Problems (TSPs), in terms of latency and energy. However, most of these methods still face unacceptable quality degradation as the problem size increases beyond a certain extent. Additionally, their hardware-agnostic adoptions limit their ability to fully exploit available hardware resources. In this w…
▽ More
Ising solvers with hierarchical clustering have shown promise for large-scale Traveling Salesman Problems (TSPs), in terms of latency and energy. However, most of these methods still face unacceptable quality degradation as the problem size increases beyond a certain extent. Additionally, their hardware-agnostic adoptions limit their ability to fully exploit available hardware resources. In this work, we introduce TAXI -- an in-memory computing-based TSP accelerator with crossbar(Xbar)-based Ising macros. Each macro independently solves a TSP sub-problem, obtained by hierarchical clustering, without the need for any off-macro data movement, leading to massive parallelism. Within the macro, Spin-Orbit-Torque (SOT) devices serve as compact energy-efficient random number generators enabling rapid "natural annealing". By leveraging hardware-algorithm co-design, TAXI offers improvements in solution quality, speed, and energy-efficiency on TSPs up to 85,900 cities (the largest TSPLIB instance). TAXI produces solutions that are only 22% and 20% longer than the Concorde solver's exact solution on 33,810 and 85,900 city TSPs, respectively. TAXI outperforms a current state-of-the-art clustering-based Ising solver, being 8x faster on average across 20 benchmark problems from TSPLib.
△ Less
Submitted 17 April, 2025;
originally announced April 2025.
-
Hybrid NER System for Multi-Source Offer Feeds
Authors:
Anusha Holla,
Bharat Gaind,
Vikas Reddy Katta,
Abhishek Kundu,
S Kamalesh
Abstract:
Data available across the web is largely unstructured. Offers published by multiple sources like banks, digital wallets, merchants, etc., are one of the most accessed advertising data in today's world. This data gets accessed by millions of people on a daily basis and is easily interpreted by humans, but since it is largely unstructured and diverse, using an algorithmic way to extract meaningful i…
▽ More
Data available across the web is largely unstructured. Offers published by multiple sources like banks, digital wallets, merchants, etc., are one of the most accessed advertising data in today's world. This data gets accessed by millions of people on a daily basis and is easily interpreted by humans, but since it is largely unstructured and diverse, using an algorithmic way to extract meaningful information out of these offers is hard. Identifying the essential offer entities (for instance, its amount, the product on which the offer is applicable, the merchant providing the offer, etc.) from these offers plays a vital role in targeting the right customers to improve sales. This work presents and evaluates various existing Named Entity Recognizer (NER) models which can identify the required entities from offer feeds. We also propose a novel Hybrid NER model constructed by two-level stacking of Conditional Random Field, Bidirectional LSTM and Spacy models at the first level and an SVM classifier at the second. The proposed hybrid model has been tested on offer feeds collected from multiple sources and has shown better performance in the offer domain when compared to the existing models.
△ Less
Submitted 11 June, 2019; v1 submitted 24 January, 2019;
originally announced January 2019.