About

Lili Su is an Assistant Professor in the Department of Electrical and Computer Engineering at Northeastern University College of Engineering, having joined in August 2020. Her research focuses on distributed machine learning, security and fault-tolerance, neural computation, bio-inspired distributed algorithms, blockchains, autonomous cars, and algorithm design. She leads the Efficient and Robust Distributed Machine Learning Laboratory, where she investigates the theoretical foundations of federated learning, including utilizing underlying data statistics to mitigate heterogeneity and client faults, supported by her NSF CAREER Award received in 2023. Dr. Su holds a PhD and MS in Electrical and Computer Engineering from the University of Illinois at Urbana-Champaign, earned in 2017 and 2014 respectively. Her notable contributions include securing distributed gradient descent in high-dimensional statistical learning, non-Bayesian learning in the presence of Byzantine adversaries, and distributed statistical machine learning in adversarial settings. Her work has been recognized with awards such as the NSF CAREER Award and Best Student Paper Awards, and she has contributed to advancing the understanding of neural computation, blockchain security, and federated learning through her research.

Research topics

• Computer Science
• Geography
• Distributed computing
• Mathematics
• Mathematical optimization
• Algorithm
• Radiology
• Medicine
• Medical physics
• Internal medicine

Selected publications

• Identification of antibiotic occurrence patterns in a human-influenced river system through dissolved organic matter analysis

  Environmental Pollution · 2025-07-30 · 4 citations

  article
  PublisherDOI

• Fast and Robust State Estimation and Tracking via Hierarchical Learning

  IEEE Transactions on Automatic Control · 2025-10-13

  articleSenior author

  Fast and reliable state estimation and tracking are essential for real-time situation awareness in Cyber-Physical Systems (CPS) operating in tactical environments or complicated civilian environments. Traditional centralized solutions do not scale well whereas existing fully distributed solutions over large networks suffer slow convergence, and are vulnerable to a wide spectrum of communication failures. In this paper, we aim to speed up the convergence and enhance the resilience of state estimation and tracking for large-scale networks using a simple hierarchical system architecture. We propose two “consensus + innovation” algorithms, both of which rely on a novel hierarchical push-sum consensus component. We characterize their convergence rates under a linear local observation model and minimal technical assumptions. We numerically validate our algorithms through simulation studies of underwater acoustic networks and large-scale synthetic networks.

  PublisherDOI

• Editor’s note to “Ultrasound image-guided cancer gene therapy using iRGD dual-targeted magnetic cationic microbubbles” [Biomed. Pharmacother. 172 (2024) 116221]

  Biomedicine & Pharmacotherapy · 2025-07-02

  articleOpen access
  PublisherDOI

• Latency-optimized multi-task collaborative computing mechanism based on NOMA-D2D for AIoT

  Computer Communications · 2025-07-04

  articleCorresponding
  PublisherDOI

• Introduction to the Special Issue on Performance Evaluation of Federated Learning Systems Part 2

  ACM Transactions on Modeling and Performance Evaluation of Computing Systems · 2025-05-30

  articleSenior author
  PublisherDOI

• Erythrabyssin ll is identified as a late-stage autophagy inhibitor reversing chemoresistance and promoting apoptosis in ovarian cancer

  iScience · 2025-06-11 · 1 citations

  articleOpen access

  Linn were screened in this study. Among them, the pterocarpan erythrabyssin II (EL-19) is a potent late-stage autophagy inhibitor, which could effectively block the fusion of autophagosome and lysosome, leading to the accumulation of autophagic substrates in both ovarian cancer A2780 and A2780/DDP cells. EL-19 did not impair the lysosomal pH and lysosomal enzyme activity. In addition, cell studies, and organoid experiments showed that EL-19 inhibited the value addition of A2780 and A2780/DDP cells, suppressed ovarian cancer organoid activity and induced apoptosis, and blocked cisplatin-induced protective autophagy in A2780/DDP cells. Combination therapy with DDP superior anti-tumor outcomes compared to monotherapies in animal models. In summary, EL-19 may be developed as an anticancer agent by blocking chemotherapy-induced protective autophagy.

  PublisherDOI

• Introduction to the Special Issue on Performance Evaluation of Federated Learning Systems Part 1

  ACM Transactions on Modeling and Performance Evaluation of Computing Systems · 2025-03-12

  articleSenior author
  PublisherDOI

• Empowering Federated Learning With Implicit Gossiping: Mitigating Connection Unreliability Amidst Unknown and Arbitrary Dynamics

  IEEE Transactions on Signal Processing · 2025-01-01 · 1 citations

  articleSenior author

  Federated learning is a popular distributed learning approach for training a machine learning model without disclosing raw data. It consists of a parameter server and a possibly large collection of clients (e.g., in cross-device federated learning) that may operate in congested and changing environments. In this paper, we study federated learning in the presence of stochastic and dynamic communication failures wherein the uplink between the parameter server and client <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$i$</tex-math></inline-formula> is on with <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">unknown</i> probability <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$p_{i}^{t}$</tex-math></inline-formula> in round <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$t$</tex-math></inline-formula>. Furthermore, we allow the dynamics of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$p_{i}^{t}$</tex-math></inline-formula> to be <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">arbitrary</i>. We first demonstrate that when the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$p_{i}^{t}$</tex-math></inline-formula>'s vary across clients, the most widely adopted federated learning algorithm, Federated Average (FedAvg), experiences significant bias. To address this observation, we propose Federated Postponed Broadcast (FedPBC), a simple variant of FedAvg. It differs from FedAvg in that the parameter server postpones broadcasting the global model to the clients with active uplinks till the end of each training round. Despite uplink failures, we show that FedPBC converges to a stationary point of the original non-convex objective. On the technical front, postponing the global model broadcasts enables implicit gossiping among the clients with active links in round <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$t$</tex-math></inline-formula>. In spite of the time-varying nature of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$p_{i}^{t}$</tex-math></inline-formula>, we can bound the perturbation of the global model dynamics using techniques to control gossip-type information mixing errors. Extensive experiments have been conducted on real-world datasets over diversified unreliable uplink patterns to corroborate our analysis.

  PublisherDOI

• Probe-Me-Not: Protecting Pre-trained Encoders from Malicious Probing

  2025-01-01

  article
  PublisherDOI

• Dynamic Aware: Adaptive Multi-Mode Out-of-Distribution Detection for Trajectory Prediction in Autonomous Vehicles

  arXiv (Cornell University) · 2025-09-16

  preprintOpen accessSenior author

  Trajectory prediction is central to the safe and seamless operation of autonomous vehicles (AVs). In deployment, however, prediction models inevitably face distribution shifts between training data and real-world conditions, where rare or underrepresented traffic scenarios induce out-of-distribution (OOD) cases. While most prior OOD detection research in AVs has concentrated on computer vision tasks such as object detection and segmentation, trajectory-level OOD detection remains largely underexplored. A recent study formulated this problem as a quickest change detection (QCD) task, providing formal guarantees on the trade-off between detection delay and false alarms [1]. Building on this foundation, we propose a new framework that introduces adaptive mechanisms to achieve robust detection in complex driving environments. Empirical analysis across multiple real-world datasets reveals that prediction errors -- even on in-distribution samples -- exhibit mode-dependent distributions that evolve over time with dataset-specific dynamics. By explicitly modeling these error modes, our method achieves substantial improvements in both detection delay and false alarm rates. Comprehensive experiments on established trajectory prediction benchmarks show that our framework significantly outperforms prior UQ- and vision-based OOD approaches in both accuracy and computational efficiency, offering a practical path toward reliable, driving-aware autonomy.

  PublisherOA PDFDOI

Frequent coauthors

• Nitin H. Vaidya

  21 shared

• Gulzira Arkin

  Jinan University

  16 shared

• Fengjuan Guo

  Jinan University

  16 shared

• Tianzhen He

  Nantong University

  12 shared

• Jinfeng Xu

  Jinan University

  12 shared

• Yingying Liu

  Southern University of Science and Technology

  12 shared

• Yao Zhu

  12 shared

• Xiaoshu Lai

  Jinan University

  12 shared

Education

• Postdoc , Computer Science & Artificial Intelligence Laboratory (CSAIL)

  Massachusetts Institute of Technology

  2020

• PhD, ECE

  University of Illinois Urbana-Champaign Grainger College of Engineering

  2017

Awards & honors

• NSF CAREER Award (2023)
• Best Student Paper Award Finalist, DISC 2016
• Best Student Paper Award, SSS 2015
• Rising Stars EECS 2018