About

Christos Faloutsos is the Fredkin Professor of Computer Science at Carnegie Mellon University, with a courtesy appointment in Electrical and Computer Engineering. He holds Ph.D. and M.Sc. degrees from the University of Toronto and a B.Sc. from the National Technical University of Athens. His research interests encompass anomaly and fraud detection in graphs and time series, human trafficking detection, fractals, self-similarity, and power laws. He is also engaged in indexing and data mining for video, biological, and medical databases. Faloutsos has contributed to a wide array of projects including stream mining, graph mining, biomedical and network data analysis using tensors, and image mining in biological contexts. His work extends to motion capture indexing, robust inter-domain routing, and interactive biological image search. He has been involved in ongoing efforts to detect human trafficking and electronic bee veterinary frameworks, as well as fraud detection in phone call and financial networks. His research has been supported by numerous NSF grants and collaborations with industry partners. Recognized for his influence in the field, Faloutsos has been featured in various media outlets and received awards such as the KDD'16 best paper award for his work on detecting fake social media followers and reviews.

Research topics

• Artificial Intelligence
• Computer Science
• Data Mining
• Computer Security
• Data science
• World Wide Web
• Database
• Human–computer interaction
• Mathematics

Selected publications

• FraudFox: Adaptable Fraud Detection in the Real World

  arXiv (Cornell University) · 2026-03-13

  preprintOpen accessSenior author

  The proposed method (FraudFox) provides solutions to adversarial attacks in a resource constrained environment. We focus on questions like the following: How suspicious is `Smith', trying to buy \$500 shoes, on Monday 3am? How to merge the risk scores, from a handful of risk-assessment modules (`oracles') in an adversarial environment? More importantly, given historical data (orders, prices, and what-happened afterwards), and business goals/restrictions, which transactions, like the `Smith' transaction above, which ones should we `pass', versus send to human investigators? The business restrictions could be: `at most $x$ investigations are feasible', or `at most \$$y$ lost due to fraud'. These are the two research problems we focus on, in this work. One approach to address the first problem (`oracle-weighting'), is by using Extended Kalman Filters with dynamic importance weights, to automatically and continuously update our weights for each 'oracle'. For the second problem, we show how to derive an optimal decision surface, and how to compute the Pareto optimal set, to allow what-if questions. An important consideration is adaptation: Fraudsters will change their behavior, according to our past decisions; thus, we need to adapt accordingly. The resulting system, \method, is scalable, adaptable to changing fraudster behavior, effective, and already in \textbf{production} at Amazon. FraudFox augments a fraud prevention sub-system and has led to significant performance gains.

  PublisherDOI

• Feedback Control for Multi-Objective Graph Self-Supervision

  ArXiv.org · 2026-02-04

  articleOpen accessSenior author

  Can multi-task self-supervised learning on graphs be coordinated without the usual tug-of-war between objectives? Graph self-supervised learning (SSL) offers a growing toolbox of pretext objectives: mutual information, reconstruction, contrastive learning; yet combining them reliably remains a challenge due to objective interference and training instability. Most multi-pretext pipelines use per-update mixing, forcing every parameter update to be a compromise, leading to three failure modes: Disagreement (conflict-induced negative transfer), Drift (nonstationary objective utility), and Drought (hidden starvation of underserved objectives). We argue that coordination is fundamentally a temporal allocation problem: deciding when each objective receives optimization budget, not merely how to weigh them. We introduce ControlG, a control-theoretic framework that recasts multi-objective graph SSL as feedback-controlled temporal allocation by estimating per-objective difficulty and pairwise antagonism, planning target budgets via a Pareto-aware log-hypervolume planner, and scheduling with a Proportional-Integral-Derivative (PID) controller. Across 9 datasets, ControlG consistently outperforms state-of-the-art baselines, while producing an auditable schedule that reveals which objectives drove learning.

  PublisherOA PDF

• AICodeDetect: A Pipeline for Systematic Detection and Analysis of AI-Generated Code

  Lecture notes in networks and systems · 2026-01-01

  book-chapterSenior author
  PublisherDOI

• Feedback Control for Multi-Objective Graph Self-Supervision

  Open MIND · 2026-02-04

  preprintSenior author

  Can multi-task self-supervised learning on graphs be coordinated without the usual tug-of-war between objectives? Graph self-supervised learning (SSL) offers a growing toolbox of pretext objectives: mutual information, reconstruction, contrastive learning; yet combining them reliably remains a challenge due to objective interference and training instability. Most multi-pretext pipelines use per-update mixing, forcing every parameter update to be a compromise, leading to three failure modes: Disagreement (conflict-induced negative transfer), Drift (nonstationary objective utility), and Drought (hidden starvation of underserved objectives). We argue that coordination is fundamentally a temporal allocation problem: deciding when each objective receives optimization budget, not merely how to weigh them. We introduce ControlG, a control-theoretic framework that recasts multi-objective graph SSL as feedback-controlled temporal allocation by estimating per-objective difficulty and pairwise antagonism, planning target budgets via a Pareto-aware log-hypervolume planner, and scheduling with a Proportional-Integral-Derivative (PID) controller. Across 9 datasets, ControlG consistently outperforms state-of-the-art baselines, while producing an auditable schedule that reveals which objectives drove learning.

  DOI

• FraudFox: Adaptable Fraud Detection in the Real World

  ArXiv.org · 2026-03-13

  articleOpen accessSenior author

  The proposed method (FraudFox) provides solutions to adversarial attacks in a resource constrained environment. We focus on questions like the following: How suspicious is `Smith', trying to buy \$500 shoes, on Monday 3am? How to merge the risk scores, from a handful of risk-assessment modules (`oracles') in an adversarial environment? More importantly, given historical data (orders, prices, and what-happened afterwards), and business goals/restrictions, which transactions, like the `Smith' transaction above, which ones should we `pass', versus send to human investigators? The business restrictions could be: `at most $x$ investigations are feasible', or `at most \$$y$ lost due to fraud'. These are the two research problems we focus on, in this work. One approach to address the first problem (`oracle-weighting'), is by using Extended Kalman Filters with dynamic importance weights, to automatically and continuously update our weights for each 'oracle'. For the second problem, we show how to derive an optimal decision surface, and how to compute the Pareto optimal set, to allow what-if questions. An important consideration is adaptation: Fraudsters will change their behavior, according to our past decisions; thus, we need to adapt accordingly. The resulting system, \method, is scalable, adaptable to changing fraudster behavior, effective, and already in \textbf{production} at Amazon. FraudFox augments a fraud prevention sub-system and has led to significant performance gains.

  PublisherOA PDF

• Understanding Transformers for Time Series: Rank Structure, Flow-of-ranks, and Compressibility

  ArXiv.org · 2025-10-02

  preprintOpen access

  Transformers are widely used across data modalities, and yet the principles distilled from text models often transfer imperfectly to models trained to other modalities. In this paper, we analyze Transformers through the lens of rank structure. Our focus is on the time series setting, where the structural properties of the data differ remarkably from those of text or vision. We show that time-series embeddings, unlike text or vision, exhibit sharply decaying singular value spectra: small patch sizes and smooth continuous mappings concentrate the data into low-rank subspaces. From this, we prove that the associated $Q/K/V$ projections admit accurate low-rank approximations, and that attention layers become compressible in proportion to the decay of the embedding spectrum. We introduce the concept of flow-of-ranks, a phenomenon by which nonlinear mixing across depth inflates the rank, explaining why early layers are most amenable to compression and why ranks grow with depth. Guided by these theoretical and empirical results, we use these insights to compress Chronos, a large time series foundation model, achieving a reduction of $65\%$ in inference time and $81\%$ in memory, without loss of accuracy. Our findings provide principled guidance for allocating width, depth, and heads in time series foundation models, and for exploiting their inherent compressibility.

  PublisherOA PDFDOI

• Featpilot: Automatic Feature Augmentation on Tabular Data

  2025-05-19

  article

  Tabular data within enterprises or open data repositories provide a huge opportunity for feature augmentation. Using these data sources to augment training data often boosts model performance, which is crucial in data-centric AutoML systems. Recent works on automatic feature augmentation have limited capabilities in utilizing useful features that cannot be joined with the base table without connecting through intermediate tables. We present Featpilot, a novel framework that explores and integrates high-quality features in tabular data for ML models. Featpilot evaluates a candidate feature from two aspects: (1) the efficacy of a join path connecting the feature to the base table and (2) the intrinsic value of a feature towards an ML task. Featpilot efficiently identifies high-quality features and their optimized join paths to augment the base table. Our experimental results show that Featpilot achieves up to a 10.27% improvement in ML model performance compared to state-of-the-art solutions across six public datasets.

  PublisherDOI

• Understanding the Implicit Biases of Design Choices for Time Series Foundation Models

  ArXiv.org · 2025-10-22

  preprintOpen access

  Time series foundation models (TSFMs) are a class of potentially powerful, general-purpose tools for time series forecasting and related temporal tasks, but their behavior is strongly shaped by subtle inductive biases in their design. Rather than developing a new model and claiming that it is better than existing TSFMs, e.g., by winning on existing well-established benchmarks, our objective is to understand how the various ``knobs'' of the training process affect model quality. Using a mix of theory and controlled empirical evaluation, we identify several design choices (patch size, embedding choice, training objective, etc.) and show how they lead to implicit biases in fundamental model properties (temporal behavior, geometric structure, how aggressively or not the model regresses to the mean, etc.); and we show how these biases can be intuitive or very counterintuitive, depending on properties of the model and data. We also illustrate in a case study on outlier handling how multiple biases can interact in complex ways; and we discuss implications of our results for learning the bitter lesson and building TSFMs.

  PublisherOA PDFDOI

• Spectro-Riemannian Graph Neural Networks

  ArXiv.org · 2025-02-01

  preprintOpen accessSenior author

  Can integrating spectral and curvature signals unlock new potential in graph representation learning? Non-Euclidean geometries, particularly Riemannian manifolds such as hyperbolic (negative curvature) and spherical (positive curvature), offer powerful inductive biases for embedding complex graph structures like scale-free, hierarchical, and cyclic patterns. Meanwhile, spectral filtering excels at processing signal variations across graphs, making it effective in homophilic and heterophilic settings. Leveraging both can significantly enhance the learned representations. To this end, we propose Spectro-Riemannian Graph Neural Networks (CUSP) - the first graph representation learning paradigm that unifies both CUrvature (geometric) and SPectral insights. CUSP is a mixed-curvature spectral GNN that learns spectral filters to optimize node embeddings in products of constant-curvature manifolds (hyperbolic, spherical, and Euclidean). Specifically, CUSP introduces three novel components: (a) Cusp Laplacian, an extension of the traditional graph Laplacian based on Ollivier-Ricci curvature, designed to capture the curvature signals better; (b) Cusp Filtering, which employs multiple Riemannian graph filters to obtain cues from various bands in the eigenspectrum; and (c) Cusp Pooling, a hierarchical attention mechanism combined with a curvature-based positional encoding to assess the relative importance of differently curved substructures in our graph. Empirical evaluation across eight homophilic and heterophilic datasets demonstrates the superiority of CUSP in node classification and link prediction tasks, with a gain of up to 5.3% over state-of-the-art models. The code is available at: https://github.com/amazon-science/cusp.

  PublisherOA PDFDOI

• Principled Mining, Forecasting, and Monitoring of Honeybee Time Series with EBV+

  ACM Transactions on Knowledge Discovery from Data · 2025-02-21 · 4 citations

  articleOpen access

  Honeybees, as natural crop pollinators, play a significant role in biodiversity and food production for human civilization. Bees actively regulate hive temperature (homeostasis) to maintain a colony’s proper functionality. Deviations from usual thermoregulation behavior due to external stressors (e.g., extreme environmental temperature, parasites, pesticide exposure) indicate an impending colony collapse. Anticipating such threats by forecasting hive temperature and finding changes in temperature patterns would allow beekeepers to take early preventive measures and avoid critical issues. In that case, how can we model bees’ thermoregulation behavior for an interpretable and effective hive monitoring system? In this article, we propose the principled Electronic Bee-Veterinarian Plus (EBV+) method based on the thermal diffusion equation and a novel “ sigmoid ” feedback-loop (P) controller for analyzing hive health with the following properties: (i) it is effective on multiple, real-world beehive time sequences (recorded and streaming), (ii) it is explainable with only a few parameters (e.g., hive health factor) that beekeepers can easily quantify and trust, (iii) it issues proactive alerts to beekeepers before any potential issue affecting homeostasis becomes detrimental, and (iv) it is scalable with a time complexity of \(O(t)\) for reconstructing and \(O(t\times m)\) for finding m cuts of a sequence with t time-ticks. Experimental results on multiple real-world time sequences showcase the potential and practical feasibility of EBV+. Our method yields accurate forecasting (up to 72% improvement in RMSE) with up to 600 times fewer parameters compared to baselines (ARX, seasonal ARX, Holt-winters, and DeepAR), as well as detects discontinuities and raises alerts that coincide with domain experts’ opinions. Moreover, EBV+ is scalable and fast, taking less than 1 minute on a stock laptop to reconstruct 2 months of sensor data.

  PublisherOA PDFDOI

Recent grants

• Finding Patterns and Anomalies in Large Time-Evolving Graphs

  NSF · $338k · 2006–2009

• III-COR: Collaborative Research: Mining Biomedical and Network Data Using Tensors

  NSF · $308k · 2007–2010

• III: Small: Influence and Virus Propagation in Large Graphs - Theory and Algorithms

  NSF · $500k · 2010–2013

• CGV: Small: Making Sense out of Large Graphs - Bridging HCI with Data Mining

  NSF · $529k · 2012–2016

• Collaborative Research: NetMine: Finding Patterns in Network Data

  NSF · $120k · 2002–2006

Frequent coauthors

• Danai Koutra

  101 shared

• Bryan Hooi

  73 shared

• Evangelos E. Papalexakis

  University of California, Riverside

  52 shared

• Rakesh Agrawal

  51 shared

• Leman Akoglu

  Carnegie Mellon University

  50 shared

• Daniel Barbará

  George Mason University

  50 shared

• Stefano Ceri

  Politecnico di Milano

  50 shared

• Christian S. Jensen

  50 shared

Labs

• Wellness NetworkPI

  Not provided

Education

• B.S.

  Nat. Tech. U. Athens

• M.S.

  University of Toronto

• Ph.D.

  University of Toronto

Awards & honors

• Ranked among the top 50 nurturers in information technology…
• Mentioned in the Greek newspaper 'To Vima' (2005)
• Distinguished Database Profiles in ACM SIGMOD record (2005)
• CMU press release about the NetProbe project for auction fra…
• NSF press release 12-187 on BIG DATA grants (2012)