About

Lav Varshney is an Associate Professor in the Electrical and Computer Engineering department at the University of Illinois, Urbana-Champaign, affiliated with the Siebel School of Computing and Data Science. His research focuses on the science and engineering of informational systems involving humans and machines, driven by a desire to improve individual and collective intelligence in modern environments. He has taught courses such as Probability with Engineering Applications, Information Theory, and Generative AI Models, reflecting his expertise in information theory and AI. His work emphasizes the intersection of AI research, public policy, and societal impact, with recent recognition for contributions to health research and AI safety. Lav Varshney is actively involved in connecting AI research with executive policy and public service, highlighting his commitment to making AI safer and less intimidating. His research and public engagement aim to advance understanding and application of AI and information systems in real-world contexts.

Research topics

• Computer Science
• Political Science
• Engineering
• Business
• Medicine
• Library science
• Agricultural economics
• Operations research
• Data science
• Engineering ethics
• Human–computer interaction
• Natural resource economics
• Marketing
• Economics
• Geography

Selected publications

• CAETC: Causal Autoencoding and Treatment Conditioning for Counterfactual Estimation over Time

  ArXiv.org · 2026-03-12

  articleOpen accessSenior author

  Counterfactual estimation over time is important in various applications, such as personalized medicine. However, time-dependent confounding bias in observational data still poses a significant challenge in achieving accurate and efficient estimation. We introduce causal autoencoding and treatment conditioning (CAETC), a novel method for this problem. Built on adversarial representation learning, our method leverages an autoencoding architecture to learn a partially invertible and treatment-invariant representation, where the outcome prediction task is cast as applying a treatment-specific conditioning on the representation. Our design is independent of the underlying sequence model and can be applied to existing architectures such as long short-term memories (LSTMs) or temporal convolution networks (TCNs). We conduct extensive experiments on synthetic, semi-synthetic, and real-world data to demonstrate that CAETC yields significant improvement in counterfactual estimation over existing methods.

  PublisherOA PDF

• Computing with Living Neurons: Chaos-Controlled Reservoir Computing with Knowledge Transplant

  arXiv (Cornell University) · 2026-04-02

  preprintOpen access

  We introduce chaos-controlled Reservoir Computing (cc-RC) for living neural cultures: dynamically rich substrates of unique potential for adaptive computation. To account for intrinsic biological variability, cc-RC combines: (i) pre-training identification of each culture's dynamical signature and phase-portrait attractor; (ii) low-power optical chaos control to stabilize spontaneous and stimulus-evoked activity; (iii) readout training within this controlled regime. Across hundreds of neural samples, cc-RC enables robust learning and pattern classification, improving both accuracy and model longevity by approximately 300% over standard RC. We further propose Knowledge Transplant (KT), for which the reservoir map learned by an expert culture is transplanted to an attractor-equivalent student culture, reducing training time to minutes while improving performance. By enabling cross-substrate, reusable learned models, KT paves the way for knowledge accumulation and sharing across neural populations, transcending biological lifespan limits.

  PublisherDOI

• Context-Gated Associative Retrieval: From Theory to Transformers

  ArXiv.org · 2026-05-08

  articleOpen accessSenior author

  Hopfield networks and their generalizations have established deep connections among biological associative memories, statistical physics, and transformers. Yet most models treat retrieval as a fixed query-to-memory mapping, ignoring the role of external context in recall. In this work, we propose a two-stage associative memory architecture, wherein a context-gate subcircuit reshapes the retrieval energy landscape before and during recall. We show theoretically that context gating increases inter-memory separation while inducing sparsity, translating into exponential improvements in retrieval. Crucially, we prove that the system admits a unique self-consistent fixed point, revealing that the resulting retrieval state is driven by both a direct contextual bias and a second-order retrieval-gate feedback loop. We then bridge this theory to transformers; specifically, we evaluate a first-order approximation on Llama-3, confirming that in-context learning acts as context-gated retrieval. Native dynamics mirror our theory: context localizes a memory subspace, enabling the zero-shot query to cleanly discriminate. Ultimately, this framework provides a mechanistic link between associative memory theory and LLM phenomenology.

  PublisherOA PDF

• Computing with Living Neurons: Chaos-Controlled Reservoir Computing with Knowledge Transplant

  ArXiv.org · 2026-04-02

  articleOpen access

  We introduce chaos-controlled Reservoir Computing (cc-RC) for living neural cultures: dynamically rich substrates of unique potential for adaptive computation. To account for intrinsic biological variability, cc-RC combines: (i) pre-training identification of each culture's dynamical signature and phase-portrait attractor; (ii) low-power optical chaos control to stabilize spontaneous and stimulus-evoked activity; (iii) readout training within this controlled regime. Across hundreds of neural samples, cc-RC enables robust learning and pattern classification, improving both accuracy and model longevity by approximately 300% over standard RC. We further propose Knowledge Transplant (KT), for which the reservoir map learned by an expert culture is transplanted to an attractor-equivalent student culture, reducing training time to minutes while improving performance. By enabling cross-substrate, reusable learned models, KT paves the way for knowledge accumulation and sharing across neural populations, transcending biological lifespan limits.

  PublisherOA PDF

• The network architecture of general intelligence in the human connectome

  Nature Communications · 2026-01-26 · 2 citations

  articleOpen access

  Advances in network neuroscience challenge the view that general intelligence (g) emerges from a primary brain region or network. Network Neuroscience Theory (NNT) proposes that g arises from coordinated activity across the brain’s global network architecture. We tested predictions from NNT in 831 healthy young adults from the Human Connectome Project. We jointly modeled the brain’s structural topology and intrinsic functional covariation patterns to capture its global topological organization. Our investigation provided evidence that g (1) engages multiple networks, supporting the principle of distributed processing; (2) relies on weak, long-range connections, emphasizing an efficient and globally coordinated network; (3) recruits regions that orchestrate network interactions, supporting the role of modal control in driving global activity; and (4) depends on a small-world architecture for system-wide communication. These results support a shift in perspective from prevailing localist models to a theory that grounds intelligence in the global topology of the human connectome. General intelligence (g) emerges from the global topology of the human connectome. Modeling structure and function in 831 adults reveals g engages distributed networks, weak long-range connections, modal control regions, and a small-world topology.

  PublisherOA PDFDOI

• Context-Gated Associative Retrieval: From Theory to Transformers

  arXiv (Cornell University) · 2026-05-08

  preprintOpen accessSenior author

  Hopfield networks and their generalizations have established deep connections among biological associative memories, statistical physics, and transformers. Yet most models treat retrieval as a fixed query-to-memory mapping, ignoring the role of external context in recall. In this work, we propose a two-stage associative memory architecture, wherein a context-gate subcircuit reshapes the retrieval energy landscape before and during recall. We show theoretically that context gating increases inter-memory separation while inducing sparsity, translating into exponential improvements in retrieval. Crucially, we prove that the system admits a unique self-consistent fixed point, revealing that the resulting retrieval state is driven by both a direct contextual bias and a second-order retrieval-gate feedback loop. We then bridge this theory to transformers; specifically, we evaluate a first-order approximation on Llama-3, confirming that in-context learning acts as context-gated retrieval. Native dynamics mirror our theory: context localizes a memory subspace, enabling the zero-shot query to cleanly discriminate. Ultimately, this framework provides a mechanistic link between associative memory theory and LLM phenomenology.

  PublisherDOI

• CAETC: Causal Autoencoding and Treatment Conditioning for Counterfactual Estimation over Time

  arXiv (Cornell University) · 2026-03-12

  preprintOpen accessSenior author

  Counterfactual estimation over time is important in various applications, such as personalized medicine. However, time-dependent confounding bias in observational data still poses a significant challenge in achieving accurate and efficient estimation. We introduce causal autoencoding and treatment conditioning (CAETC), a novel method for this problem. Built on adversarial representation learning, our method leverages an autoencoding architecture to learn a partially invertible and treatment-invariant representation, where the outcome prediction task is cast as applying a treatment-specific conditioning on the representation. Our design is independent of the underlying sequence model and can be applied to existing architectures such as long short-term memories (LSTMs) or temporal convolution networks (TCNs). We conduct extensive experiments on synthetic, semi-synthetic, and real-world data to demonstrate that CAETC yields significant improvement in counterfactual estimation over existing methods.

  PublisherDOI

• Energy-Aware Routing to Large Reasoning Models

  IEEE Journal on Selected Areas in Information Theory · 2026-01-01

  articleOpen accessSenior author
  PublisherDOI

• Learning from one and only one shot

  npj Artificial Intelligence · 2025-07-14 · 4 citations

  articleOpen access

  Abstract Humans can generalize from only a few examples and from little pretraining on similar tasks. Yet, machine learning (ML) typically requires large data to learn or pre-learn to transfer. Motivated by nativism and artificial general intelligence, we directly model human-innate priors in abstract visual tasks such as character and doodle recognition. This yields a white-box model that learns general-appearance similarity by mimicking how humans naturally “distort” an object at first sight. Using just nearest-neighbor classification on this cognitively-inspired similarity space, we achieve human-level recognition with only 1–10 examples per class and no pretraining. This differs from few-shot learning using massive pretraining. In the only-few-shot regime of MNIST, EMNIST, Omniglot, and QuickDraw benchmarks, we outperform both modern neural networks and classical ML. For unsupervised learning, by learning the non-Euclidean, general-appearance similarity space in a k -means style, we achieve multifarious visual realizations of abstract concepts by generating human-intuitive archetypes as cluster centroids.

  PublisherOA PDFDOI

• Building sustainable and resilient agri-food systems under multiple shocks

  Frontiers in Sustainable Food Systems · 2025-12-08 · 1 citations

  articleOpen access

  Shocks, such as disease outbreaks, extreme weather events, cyberattacks, financial crises, and wars, are occurring with greater frequency. When these shocks occur simultaneously and/or in sequence, referred to here as multiple shocks, they can generate compound impacts on agri-food systems and contribute to food and nutrition insecurity. Building sustainable agri-food systems that are resilient to multiple shocks requires an integrated understanding of the threats posed by multiple shocks to all aspects of supply chain networks. Collective action by researchers, educators, extension experts, and other stakeholders can mitigate and improve adaptation to these impacts. However, there are major knowledge gaps in examining, understanding, and synthesizing agri-food systems under multiple shocks. Previous actions have been fragmented, as efforts have largely focused only on an individual shock, in a specific place, and with separate rather than integrated efforts in research, education, and extension. Here, we present an integrated framework to address multiple shocks toward enhancing agri-food system resilience and sustainability. We illustrate how this integrated framework can be operationalized, focusing on assessing impacts, identifying mitigation strategies, providing decision support, training a future agri-food system workforce, and building communities for resilience to multiple shocks. Finally, we discuss challenges and opportunities in applying the framework for enhancing agri-food system resilience and sustainability worldwide, thus contributing to the realization of several United Nations Sustainable Development Goals, particularly SDG 2 (Zero Hunger).

  PublisherOA PDFDOI

Recent grants

• CIF: Small: Foundations of Belief Sharing in Human-Machine Systems

  NSF · $405k · 2017–2022

• CIF: EAGER: Towards an Information Theory of Attention

  NSF · $175k · 2016–2019

Frequent coauthors

• Ivan Brugere

  46 shared

• Thomas Yu

  43 shared

• Jifan Gao

  University of Wisconsin–Madison

  43 shared

• Sean D. Mooney

  National Institutes of Health

  43 shared

• Guanhua Chen

  University of Hong Kong

  43 shared

• Zofia Nawalany

  42 shared

• Augustinas Prusokas

  Newcastle University

  42 shared

• Sunkyu Kim

  Incyte (United States)

  42 shared

Education

• Ph.D., Computer Science

  University of Illinois at Urbana-Champaign

  2000

• M.S., Computer Science

  University of Illinois at Urbana-Champaign

  1996

• B.S., Computer Science

  University of Illinois at Urbana-Champaign

  1994

Awards & honors

• Celebration of Excellence 2023
• Celebration of Excellence 2022
• Celebration of Excellence 2021
• Celebration of Excellence 2024
• Celebration of Excellence 2025