About

Nikil Dutt is a Chancellor’s Professor at the University of California, Irvine, with academic appointments in the departments of Computer Science, Electrical Engineering and Computer Science, and Cognitive Sciences. He is affiliated with multiple research centers at UCI, including the Center for Embedded Computer Systems, the Center for Cognitive Neuroscience and Engineering, the California Institute for Telecommunications and Information Technology, the Center for Pervasive Communications and Computing, and the Laboratory for Ubiquitous Computing and Interaction. His research interests encompass embedded systems, electronic design automation, computer architecture, optimizing compilers, system specification techniques, distributed systems, formal methods, and brain-inspired architectures and computing.

Research topics

• Medicine
• Computer Science
• Psychiatry
• Psychology
• Physical therapy
• Internal medicine
• Sociology
• Nursing
• Political Science
• Gerontology
• Clinical psychology
• Data science
• Engineering
• Knowledge management
• Surgery
• Embedded system

Selected publications

• MedSpeak: A Knowledge Graph-Aided ASR Error Correction Framework for Spoken Medical QA

  2026-04-21

  article
  PublisherDOI

• HYPERDOA: Robust and Efficient DoA Estimation Using Hyperdimensional Computing

  2026-04-21

  articleSenior author

  Direction of Arrival (DoA) estimation techniques face a critical trade-off, as classical methods often lack accuracy in challenging, low signal-to-noise ratio (SNR) conditions, while modern deep learning approaches are too energy-intensive and opaque for resource-constrained, safety-critical systems. We introduce HYPERDOA, a novel estimator leveraging Hyperdimensional Computing (HDC). The framework introduces two distinct feature extraction strategies— Mean Spatial-Lag Autocorrelation and Spatial Smoothing—for its HDC pipeline, and then reframes DoA estimation as a pattern recognition problem. This approach leverages HDC’s inherent robustness to noise and its transparent algebraic operations to bypass the expensive matrix decompositions and "black-box" nature of classical and deep learning methods, respectively. Our evaluation demonstrates that HYPERDOA achieves ∼35.39% higher accuracy than state-of-the-art methods in low-SNR, coherent-source scenarios. Crucially, it also consumes ~93% less energy than competing neural baselines on an embedded NVIDIA Jetson Xavier NX platform. This dual advantage in accuracy and efficiency establishes HYPERDOA as a robust and viable solution for mission-critical applications on edge devices.

  PublisherDOI

• GOLD: Green Optimization of Language Models Serving on Devices

  2025-01-11 · 1 citations

  articleSenior author

  With the growing integration of language models into embedded devices, enhancing energy efficiency while maintaining target performance is essential. This paper presents GOLD, a policy that enhances on-device language model inference using DVFS and power management via reinforcement learning. GOLD optimizes energy use, extending battery life by up to <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$1.39\times$</tex> and increasing performance by 4%. This solution supports sustainable AI for energy-limited environments, meeting the rising demand for LLM on the device.

  PublisherDOI

• ACCESS-AV: Adaptive Communication-Computation Codesign for Sustainable Autonomous Vehicle Localization in Smart Factories

  ArXiv.org · 2025-07-27

  preprintOpen accessSenior author

  Autonomous Delivery Vehicles (ADVs) are increasingly used for transporting goods in 5G network-enabled smart factories, with the compute-intensive localization module presenting a significant opportunity for optimization. We propose ACCESS-AV, an energy-efficient Vehicle-to-Infrastructure (V2I) localization framework that leverages existing 5G infrastructure in smart factory environments. By opportunistically accessing the periodically broadcast 5G Synchronization Signal Blocks (SSBs) for localization, ACCESS-AV obviates the need for dedicated Roadside Units (RSUs) or additional onboard sensors to achieve energy efficiency as well as cost reduction. We implement an Angle-of-Arrival (AoA)-based estimation method using the Multiple Signal Classification (MUSIC) algorithm, optimized for resource-constrained ADV platforms through an adaptive communication-computation strategy that dynamically balances energy consumption with localization accuracy based on environmental conditions such as Signal-to-Noise Ratio (SNR) and vehicle velocity. Experimental results demonstrate that ACCESS-AV achieves an average energy reduction of 43.09% compared to non-adaptive systems employing AoA algorithms such as vanilla MUSIC, ESPRIT, and Root-MUSIC. It maintains sub-30 cm localization accuracy while also delivering substantial reductions in infrastructure and operational costs, establishing its viability for sustainable smart factory environments.

  PublisherOA PDFDOI

• An integrated toolbox for creating neuromorphic edge applications

  Neuromorphic Computing and Engineering · 2025-01-22 · 3 citations

  articleOpen access

  Abstract spiking neural networks (SNNs) and neuromorphic models are believed to be more efficient in general and have more biological realism than the activation functions typically used in deep neural networks, transformer models and generative AI. SNNs have local learning rules, are able to learn on small data sets, and can adapt through neuromodulation. Although research has shown their advantages, there are still few compelling practical applications, especially at the edge where sensors and actuators need to be processed in a timely fashion. One reason for this might be that SNNs are much more challenging to understand, build, and operate due to their intrinsic properties. For instance, the mathematical foundation involves differential equations rather than basic activation functions. To address these challenges, we have developed CARLsim++, an integrated toolbox that facilitates the creation of neuromorphic applications. It extends the highly efficient CARLsim open-source SNN simulator. CARLsim++ encapsulates the mathematical intrinsics and low-level C++ programming by providing a graphical user interface for users to easily create their SNNs and a means to configure sensors and actuators for robotics and other edge devices. These can be accurately simulated before deploying on physical devices. CARLsim++ can lead to rapid development of neuromorphic applications for simulation or edge processing. We introduce CARLsim++ with a closed loop robotic demonstration using neuromorphic computing.

  PublisherOA PDFDOI

• Invited Paper: Mindful AI for Pervasive Health and Wellbeing (PHW)

  2025-10-26

  articleSenior author

  Emerging AI-driven pervasive health and wellbeing (PHW) services (e.g., personalized health assistants and mobile health applications) face critical challenges in handling noisy/intermittent sensory data, integrating cross-modal insights, and stringent energy and compute constraints. We present Mindful AI, a cognitive-inspired framework designed to enable adaptive, resilient, and efficient PHW services in real-world conditions. Our dual-mode intelligence—Automatic (System 1) and Reflective (System 2)—selectively directs system attention toward the most relevant sensing and compute contexts, unifying bottom-up stimuli (driven by input quality, inference demands and model confidence, and resource availability) with top-down insights (reflecting user demands, system goals/constraints, and contextual information). Our framework distills and orchestrates insights across sensing, communication, and computation through hybrid attention toward bottom-up and top-down insights that support cross-layer sense-compute co-optimization to achieve resilient, low-latency, and energy-efficient PHW services. We evaluate our approach on multi-tier device-edge-cloud platforms, using real-world case studies in pain assessment, stress monitoring, and human activity recognition to demonstrate adaptation to real-world uncertainties (e.g., sensor degradation, context drift, network variability), while maintaining strict QoS, accuracy, and latency guarantees.

  PublisherDOI

• Feasibility, Acceptability, and Preliminary Outcomes of a Mobile Adaptation of a Relational Savoring Intervention to Prevent Loneliness in College Students: Mixed Methods Pilot Study

  JMIR Formative Research · 2025-05-28 · 4 citations

  articleOpen access

  Background: Rates of loneliness have risen sharply since the onset of the COVID-19 pandemic, largely due to disruptions in social relationships and daily routines, with college students experiencing some of the greatest increases. While prevention programs targeting loneliness have been developed, their success has been limited. One promising approach may lie in enhancing the quality of existing relationships rather than simply increasing social interactions during periods of acute loneliness. Relational savoring, an intervention rooted in attachment theory and positive psychology, aims to deepen feelings of connection by encouraging individuals to reflect on positive interpersonal experiences. Objective: This study aimed to evaluate the feasibility, acceptability, and preliminary outcomes of a mobile health adaptation of relational savoring, termed mSavorUs (developed by Amir Rahmani), designed to prevent and reduce loneliness among college students. Methods: A randomized controlled pilot study was conducted with a diverse sample of 29 college students (43.3% Latinx, 40% Asian American, and 16.7% White). The intervention leveraged a smart ring, smartwatch, and smartphone app to enable just-in-time delivery of relational savoring prompts, alongside continuous monitoring of loneliness-related indicators (eg, physiological activity, sleep, and behavior). Aim 1 involved a thematic analysis of participant feedback regarding the utility, benefits, and challenges of both mSavorUs and the monitoring tools. Aim 2 examined the intervention's effects on loneliness and perceived connectedness. Results: For aim 1, qualitative findings suggested that participants found the content of mSavorUs (developed by Amir Rahmani) rewarding and helpful; however, the timing of the intervention was often experienced as disruptive. For aim 2, quantitative analyses revealed no significant reductions in loneliness or increases in connectedness, indicating the need for adjustments to the intervention delivery method. Conclusions: Although participants found the intervention content valuable, the just-in-time delivery format may have limited its effectiveness. Future iterations should consider alternative timing or delivery strategies to maximize program benefits.

  PublisherDOI

• AAMLA: An Autonomous Agentic Framework for Memory-Aware LLM-Aided Hardware Generation

  2025-07-31

  preprintOpen accessSenior author

  Large Language Models (LLMs) have recently emerged as powerful assistants for hardware design, translating natural-language specifications into Hardware Description Languages (HDLs), yet fine-tuning these models on domainspecific corpora routinely exceeds the memory capacity of commodity GPUs and triggers Out-of-Memory (OoM) errors. We present AAMLA, an autonomous agentic framework that converts this pain point into a pushbutton experience. AAMLA incorporates memory awareness by coupling (i) a predictive memory profiler-LLMem++-that significantly extends the original LLMem framework to support a diverse set of memoryefficient fine-tuning strategies, including adapter-based (LoRA, DoRA), gradient-free (MeZO), token-sparse (TokenTune), and optimizer-modified (APOLLO) methods, with (ii) a portfolio of complementary memory-efficient adaptation techniques-leading to a complete synthesis flow. The system allows for a twostep early design space exploration workflow: it first prunes any method whose predicted footprint would violate the user's GPU budget, then consults an offline accuracy-latency atlas to recommend the Pareto-optimal strategy that aligns with the designer's stated priority (accuracy or turnaround time). Guided by this workflow, an agentic controller configures and applies the chosen technique, guaranteeing OoM-free execution without manual trial-and-error. The framework is model-and datasetagnostic, easily extensible with new tuning primitives, and exposes a simple interface that accepts natural-language prompts and emits synthesizable Verilog, thereby lowering the barrier to LLMassisted hardware design for researchers and organizations with limited computational resources.

  PublisherDOI

• Exploiting Approximation for Run-time Resource Management of Embedded HMPs

  ACM Transactions on Embedded Computing Systems · 2025-03-13 · 1 citations

  articleOpen access

  Run-time resource management (RTM) of multi-programmed workloads on heterogeneous multi-core platforms is challenging due to (i) fixed power budget of the device, (ii) variable performance requirements of the workloads, and (iii) unknown arrival of the applications. Existing RTM solutions lack power-performance coordination, resulting in performance degradation during power actuation or power violations during performance provisioning. Exploiting inherent error-resilience of the applications can address the performance loss incurred in power actuation, by combining run-time approximation with traditional power knobs (including Dynamic Voltage/Frequency Scaling, Task Migration, Degree of Parallelism, and CPU Quota ). In this work, we present an accuracy-aware resource management framework that jointly actuates run-time approximation and traditional power knobs for efficient power-performance management of multi-programmed and multi-threaded workloads running on heterogeneous mobile platforms. Our strategy configures the accuracy of the applications at run-time to exploit accuracy-performance trade-offs, by considering system-wide power-performance dynamics. We use heuristic estimation models to jointly enforce accuracy configuration and traditional power knobs settings at run-time. We evaluated our framework on real-world embedded mobile platforms, including Odroid XU3 and Asus Tinker Edge R boards to demonstrate the efficiency of our proposed approach across multiple workload scenarios. Our approach achieved 25% lower performance violations against the state-of-the-art run-time resource management policies at the cost of 2.2% accuracy loss across six applications.

  PublisherOA PDFDOI

• Exploiting Approximate SRAM for Energy-Efficient Integer Motion Estimation on VVC Encoders

  2025-05-25

  article

  Video coding is a critical technology for enabling many modern applications. However, the high complexity of state-of-the-art encoders leads to energy consumption challenges, particularly in memory systems. This paper proposes an integer motion estimation (IME) system using approximate SRAM memories to enhance the energy efficiency of VVC encoders. The approximate SRAM memories are employed for both the current block and the search area buffers by reducing the supply voltage. The proposed IME system is evaluated using a customized tool that simulates the approximation effects on both buffers based on real energy measurements from a 28nm SRAM. This tool is integrated with VVenC, a fast VVC implementation, to assess the impact on coding efficiency and energy consumption. Experimental results show that the IME system using approximate SRAM can reduce energy consumption in reading operations by up to 55%, with a low impact on coding efficiency.

  PublisherDOI

Recent grants

• Cross-Layer Error Exploitation for Next Generation SoCs

  NSF · $500k · 2007–2011

• WiFiUS: IoCT-CARE: Internet of Cognitive Things for Personalized Healthcare

  NSF · $300k · 2017–2020

• Collaborative Research: Platform-Based CAD for Power and Performance Optimization

  NSF · $96k · 2002–2006

• SHF: Medium: Conquering MPSoC Complexity with Principles of a Self-Aware Information Processing Factory

  NSF · $1.1M · 2017–2023

• SCC: UNITE: Smart, Connected, and Coordinated Maternal Care for Underserved Communities

  NSF · $2.1M · 2018–2023

Frequent coauthors

• Amir M. Rahmani

  116 shared

• Amir M. Rahmani

  Kurdistan University of Medical Sciences

  112 shared

• Fadi Kurdahi

  University of California, Irvine

  102 shared

• Alex Nicolau

  University of California, Irvine

  101 shared

• Sudeep Pasricha

  67 shared

• Prabhat Mishra

  55 shared

• Jun Yong Shin

  University of California, Irvine

  54 shared

• Sina Labbaf

  52 shared

Education

• Ph.D., Computer Science

  University of California, Los Angeles

  1993

• M.S., Computer Science

  University of California, Los Angeles

  1989

• B.S., Electrical Engineering

  University of California, Los Angeles

  1987

Awards & honors

• Best Paper Awards at CHDL’89, CHDL’91, VLSI Design 2003, COD…
• Best Paper Award Nominations at WASP 2004, DAC 2005, VLSI De…
• ACM SIGDA Distinguished Lecturer (2001-2002)
• IEEE Computer Society Distinguished Visitor (2003-2005)
• Fellow of the IEEE