Archit Gajjar

Archit Gajjar

Auto-generated profile, based on publicly available information

Archit Gajjar

Artificial Intelligence Research Lab (AIRL), Hewlett Packard Labs, USA

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 November 2024 in IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
Verified
Journal Article
Open access
GPGiacomo Pedretti
Giacomo Pedretti
JMJohn Moon
John Moon
PBPedro Bruel
Pedro Bruel
SSSergey Serebryakov
Sergey Serebryakov
RRRon M. Roth
Ron M. Roth
CGCatherine E. Graves
Catherine E. Graves
Structured, or tabular, data is the most common format in data science. While deep learning models have proven formidable in learning from unstructured data such as images or speech, they are less accurate than simpler approaches when learning from tabular data. In contrast, modern tree-based Machine Learning (ML) models shine in extracting relevant information from structured data. An essential requirement in data science is to reduce model inference latency in cases where, for example, models are used in a closed loop with simulation to accelerate scientific discovery. However, the hardware acceleration community has mostly focused on deep neural networks and largely ignored other forms of machine learning. Previous work has described the use of an analog content addressable memory (CAM) component for efficiently mapping random forests. In this work, we develop an analog-digital architecture that implements a novel increased precision analog CAM and a programmable chip for inference of state-of-the-art tree-based ML models, such as XGBoost, CatBoost, and others. Thanks to hardware-aware training, X-TIME reaches state-of-the-art accuracy and 119× higher throughput at 9740× lower latency with > 150× improved energy efficiency compared with a state-of-the-art GPU for models with up to 4096 trees and depth of 8, with a 19W peak power consumption.

18 November 2024 in ACM Transactions on Reconfigurable Technology and Systems
Verified
Journal Article
Open access
AGArchit Gajjar
Archit Gajjar
PKPriyank Kashyap
Priyank Kashyap
Aydin AysuAydin Aysu
Aydin Aysu
PFPaul Franzon
Paul Franzon
YCYongjin Choi
Yongjin Choi
JIJim Ignowski
Jim Ignowski
Over the last decade, there has been a rise in cyberattacks, particularly ransomware, causing significant disruption and financial repercussions across public and private sectors. Tremendous efforts have been spent on developing techniques to detect ransomware to, ideally, protect data or have as minimum data loss as possible. Ransomware attacks are becoming more frequent and sophisticated as there is a constant tussle between attackers and cybersecurity defenders. Machine Learning (ML) approaches have proven more effective in detecting ransomware than classical signature-based detection. In particular, tree-based algorithms such as Decision Trees (DT), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost) spike up interest among cybersecurity researchers. However, due to the nature of the problem, traditional CPUs and GPUs fail to keep up with the desired performance, especially for large data workloads. Thus, the problem demands a customized solution to detect the ransomware. Here, we propose an FPGA accelerated tree-based ML model for multi-dataset ransomware detection. We show the capability of the proposed prototype to address the problem from more than one set of features, reducing false positive and negative rates to have robust predictions by looking at Hardware Performance Counters (HPCs), Operating System (OS) calls, and network traffic information simultaneously. With 1,000 samples per batch, the FPGA prototype has 65.8 \({\times}\) and 4.1 \({\times}\) lower latency over the CPU and GPU, respectively. Moreover, the FPGA design is up to 11.3 \({\times}\) cost-effective and 643 \({\times}\) energy-efficient compared to the CPU and 3 \({\times}\) cost-effective and 16.8 \({\times}\) energy-efficient over the GPU.

1 December 2018
Verified
Conference Paper
Closed access
AGArchit Gajjar
Archit Gajjar
XYXiaokun Yang
Xiaokun Yang
HKHakduran Koc
Hakduran Koc
IUIshaq Unwala
Ishaq Unwala
LWLei Wu
Lei Wu
JLJiang Lu
Jiang Lu
This paper presents an Internet-of-Things (IoT) architecture by integrating a Synology cloud server, edge computing systems, and physical networks. More specifically, we have established a physical network combining two subsystems 1) non-standardized Bluetooth Low Energy (BLE) Mesh network, and 2) a security monitoring system. The BLE Mesh system has one IoT host connected with three BLE devices, enabling to extend the communication distance by using one or two relays. The monitoring system consists of a Passive Infrared Sensor (PIR) and a webcam with multiple solutions for recognizing a human face. Two algorithms, Low Binary Pattern Histogram (LBPH) and Deep Metric Learning (DML), have been implemented and evaluated on different benchmarks. Experimental results show that the DML-based computation can reach 99.38% accuracy with almost 400 ms latency for recognizing a single face in frames of images. The future work will focus on testing the cloud service by integrating a Synology D218+ server, as well as improving the computation speed of facial recognition on pure hardware design on field-programmable gate array (FPGA). The aim of our work is to provide a robust IoT-Edge-Cloud system which can be deployed on the large-scale applications and processes data much faster compared to traditional cloud computing system due to the perks of parallel computing on FPGAs at the network edge.

27 February 2025
Verified
Conference Paper
Open access
AGArchit Gajjar
Archit Gajjar
LZLei Zhao
Lei Zhao
OEOmar Eldash
Omar Eldash
ANAishwarya Natarajan
Aishwarya Natarajan
RJRand Jean
Rand Jean
LBLuca Buonanno
Luca Buonanno
While the AI computing needs are ever-increasing and the innovation in models generates tens of new architectures yearly, the performance gain from improvements in chip manufacturing has slowed down. Within this context, FPGA-based accelerators play a fundamental role. FPGAs are the backbone of specialized architectures, their reconfigurability being the key differentiation that enables an effective design space exploration. At the same time, to overcome the limitations induced by the memory bottleneck, the computing architectures community has proposed the in-memory computing paradigm: storage and computations are both performed in non-volatile memory devices. We propose the enhancement of FPGAs with in-memory computing macros, specifically analog Dot Product Engines, based on non-volatile ReRAM devices. We designed a novel 40\:nm,\break 23.7\:mm×23.7\:mm architecture using the VTR framework and used a custom event-driven simulator to estimate its projected performances. When comparing its performances to the state-of-the-art, the in-memory computing enhanced FPGA achieves 154\:TOPS, a ×234 throughput improvement, and 2.08\:TOPS/W, a ×143 energy efficiency improvement with respect to published results for FPGA implementations of AI models. This paper intends to demonstrate to the FPGA community the breakthrough advantages in energy efficiency and throughput derived from the integration of Dot Product Engines in the FPGA fabric, aiming at fulfilling the exponentially increasing computing requirements of the AI hardware and HPC industry.

27 March 2026 in ACM Transactions on Architecture and Code Optimization
Verified
Journal Article
Open access
AGArchit Gajjar
Archit Gajjar
RSRuthwik Sunketa
Ruthwik Sunketa
LZLei Zhao
Lei Zhao
OEOmar Eldash
Omar Eldash
ANAishwarya Natarajan
Aishwarya Natarajan
LBLuca Buonanno
Luca Buonanno
Modern AI models place heavy demands on compute resources, underscoring the importance of hardware accelerators that can balance performance, energy, and flexibility. The ever-growing demand for AI computing, coupled with slowing performance gains in chip manufacturing, has heightened the role of FPGA-based accelerators due to their rapid adaptability, reprogrammability, and support for custom parallel data flows. In this work, we introduce a domain-optimized FPGA architecture designed for deep neural network (DNN) inference by embedding analog in-memory compute blocks, specifically, RRAM-based Dot Product Engines (DPE), directly into the fabric. These engines perform multiply-accumulate operations within an RRAM-based crossbar, minimizing data movement through the FPGA routing fabric, while enhancing the compute throughput of the FPGA significantly, overall improving energy efficiency of mapping DNNs on FPGAs. We evaluate our architecture using the Verilog-to-Routing (VTR) framework. We simulate a novel 22 nm architecture, and employ a custom event-driven simulator to evaluate its performance. Our design achieves an average 6.58× reduction in latency, 5.42× throughput improvement, and 8,741× energy efficiency improvement when compared to state-of-the-art FPGA implementations across multiple DNNs (LeNet, ResNet, VGG).

28 May 2026 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
LZLei Zhao
Lei Zhao
LBLuca Buonanno
Luca Buonanno
AGArchit Gajjar
Archit Gajjar
JMJohn Moon
John Moon
ANAishwarya Natarajan
Aishwarya Natarajan
GPGiacomo Pedretti
Giacomo Pedretti
Resistive Random Access Memory (RRAM) based in-memory computing (IMC) accelerators offer significant performance and energy advantages for deep neural networks (DNNs), but face three major limitations: (1) they support only static dot-product operations and cannot accelerate arbitrary non-linear functions or data-dependent multiplications essential to modern LLMs; (2) they demand large, power-hungry analog-to-digital converter (ADC) circuits; and (3) mapping model weights to device conductance introduces errors from cell nonidealities. These challenges hinder scalable and accurate IMC acceleration as models grow. We propose NL-DPE, a Non-Linear Dot Product Engine that overcomes these barriers. NL-DPE augments crosspoint arrays with RRAM-based Analog Content Addressable Memory (ACAM) to execute arbitrary non-linear functions and data-dependent matrix multiplications in the analog domain by transforming them into decision trees, fully eliminating ADCs. To address device noise, NL-DPE uses software-based Noise Aware Fine-tuning (NAF), requiring no in-device calibration. Experiments show that NL-DPE delivers 28× energy efficiency and 249× speedup over a GPU baseline, and 22× energy efficiency and 245× speedup over existing IMC accelerators, while maintaining high accuracy.