About me

I'm a Scientist and Principal Investigator at the Max Planck Institute for Neurobiology of Behavior. One aspect of my work involves characterizing the diversity of cortical inhibitory neurons, while another focuses on exploring human-AI teaming approaches for neuroscience.

Originally trained as an Electrical Engineer, I then specialized in Machine Learning. I worked in software engineering for a couple of years before grad school. During my undergrad studies, I did internships at The ABB Group, MIT, and Inria Paris-Saclay.

Publications Communication Email

Unraveling structural principles of the cortex

Cortical inhibitory neurons display enormous diversity in their cellular properties. How do these properties relate to one another?

My approach to cast this question is as follows:

  1. Dataset creation. I standardized and curated a large representative sample of cortical inhibitory neurons.

  2. Robust classification. I assigned neurons into classes based on a highly accurate multi-modal assessment.

  3. Flexible model development. I designed a self-supervised model to learn the statistics of a representative dataset across cortical layers.

Surprisingly, depth-dependent gradients of cellular properties provide a link between them, suggesting an adjustment of cellular properties once integrated into local cortical circuits.

Human-AI collaboration for discovery neuroscience

As AI systems surpass human experts in predicting the outcomes of neuroscience experiments. What would then be the role for humans in the overall decision process?

Effective team complementarity requires:

  1. Calibrated confidence. This implies that when an agent (human or AI) is more confident in their judgments, those judgments tend to be more accurate.

  2. Classification diversity among agents. Diversity holds when the errors in classification made by humans and AIs are not the same.

Human judgment that cannot be supplanted by AI, this opens the possibility for humans and AI systems to team together to make discoveries.

Publications

Confidence-weighted integration of human and machine judgments for superior decision-making
Felipe Yáñez, Xiaoliang Luo, Omar Valerio Minero, and Bradley C. Love
Working paper
preprint code

FOXP2-immunoreactive, corticothalamic pyramidal cells in neocortical layers 6a and 6b are tightly regulated by neuromodulatory systems
Guanxiao Qi, Danqing Yang, Fernando Messore, Arco Bast, Felipe Yáñez, Marcel Oberlaender, and Dirk Feldmeyer
iScience 2025
paper preprint

Large language models surpass human experts in predicting neuroscience results
Xiaoliang Luo*, Akilles Rechardt*, Guangzhi Sun*, Kevin K. Nejad*, Felipe Yáñez*, Bati Yilmaz*, ..., and Bradley C. Love*
Nature Human Behaviour 2024
paper preprint code website
* indicates major contributions

Simulation-based inference for efficient identification of generative models in computational connectomics
Jan Boelts, Philipp Harth, Richard Gao, Daniel Udvary, Felipe Yáñez, Daniel Baum, Hans-Christian Hege, Marcel Oberlaender, and Jakob H. Macke
PLOS Computational Biology 2023
paper preprint code

A hidden parameter network that explains topological properties observed in neocortex
Felipe Yáñez, Daniel Udvary, and Marcel Oberlaender
ICLR Generalization Workshop 2021
paper poster

Primal-Dual Algorithms for Non-negative Matrix Factorization with the Kullback-Leibler Divergence
Felipe Yáñez, and Francis Bach
IEEE ICASSP 2017 (oral presentation)
paper slides preprint code

Quantitative Susceptibility Map Reconstruction via a Total Generalized Variation regularization
Felipe Yáñez, Audrey Fan, Berkin Bilgic, Carlos Milovic, Elfar Adalsteinsson, and Pablo Irarrázaval
International Workshop on Pattern Recognition in Neuroimaging 2013
paper

Teaching & outreach

  • LatinX in AI & Deep Learning Indaba

    As a mentor, I actively share my skills and experiences with early career researchers from the Latin American and African AI communities.

  • Wilbe Science

    As a BSF fellow, I bounce ideas, contribute to shared resources, and interact with an exceptional community of science entrepreneurs.

  • PUC Chile & Uni Frankfurt

    As a TA and lecturer, I deliver a wide range of courses including Machine Learning, Pattern Recognition, Computational Neuroscience, Imaging, MRI Laboratory, Fluid Mechanics, and Calculus.

    Sample materials of other courses:

    Differential Equations Signals & Systems Linear Algebra

  • Volta UC

    As an electrical engineering undergrad, I helped developing a 60kW electric vehicle and producing the first technological documentary made in Chile.


Selected presentations & talks

La Métropole

At Cosyne 2025, I presented how the structural organization of inhibitory neurons is preserved from caudal to rostral cortical areas, as well as in mouse, rat, and human. There are some interesting connectivity patterns when analyzing millimeter-scale volumetric electron microscopy data. The next step is to relate the structure to in vivo activity function. I'm grateful to the many people who stopped by to discuss such a relevant topic.

Dissertation!

I discovered that depth-dependent variations in morphoelectric properties reveal the molecular identity of inhibitory neurons. This allows for the disentanglement of variability due to intrinsic (i.e., genetic) and extrinsic (i.e., developmental) factors.


Public Lecture


What kind of humanity do we want to be?

I discussed team collaboration with superintelligent AI systems. I'm thankful to the Chilean Senate for making this possible.


City by the Lake

I gave a talk at Northwestern Feinberg School of Medicine and presented at Neuroscience 2024. I showed that the anatomical and physiological diversity of inhibitory neurons within a molecular class does not primarily reflect genetic differences, but rather arises to a large extent due to structural and functional plasticity during the development of the neocortex.


Details

Online talk

I was invited to give an online talk at McGovern. The presentation was on how depth-dependent variations in anatomical and physiological properties reveal the molecular identity of cortical inhibitory neurons. This indicates that inhibitory neurons adjust their cellular properties to local cortical circuits during development.


Hopkins and DC

I headed to Johns Hopkins University to give a talk on some very surprising observations on the anatomical and physiological organization of cortical interneurons. Then, I presented those observations at Neuroscience 2023.


Details


Rehacking time in Granada

After learning about NWB and the DANDI Archive, I reanalyzed published data to explore anatiomical and physiological variations of inhibitory interneurons across different species and cortex areas. I'm thankful to The Kavli Foundation for funding my travel and accommodation.


Hacking neuro data!

I attended the NWB users and developers meeting in July 2023. It was an incredible experience where I had the chance to work on the standarization of intracellular electrophysiology and start getting involved in the integration of LLMs that assist in the identification of dandisets relevant to a scientific question. I'm thankful to Janelia for covering all the costs of my visit including flight.


Mission KI

On an event centered around science and technology, I presented my perspective on scaling principles of the cerebral cortex. It was an excellent instance to chat with folks about challenges and opportunities, as the landscape of scientific research is transforming.


Details (in german)


Neuroscience in sunny San Diego

In November 2022, I gave a talk about robust cell type prediction of in vivo cortical inhibitory interneurons at the XXXV Barrels meeting. Then, at Neuroscience 2022, I presented a highly accurate classification of a large representative sample of cortical inhibitory interneurons.


Details

Back to Paris

After a long while I had the chance to meet with old friends, colleagues, and mentors at the FENS Forum 2022 in Paris. I presented a procedure to robustly cluster a large representative sample of cortical interneurons. I also showed that my methodology can be readily applied to multi-modal in vivo predictions and cross-species analysis of cell types.


Details

Talk @EPFL

During the summer of 2022, I gave a talk at EPFL about organizing principles of cortical interneurons.



Interview with Mel Nuesch (in Spanish)

I was invited to talk about my research on the development of robust technologies to understand structural properties of the brain.



Virtual presentation


Northwestern + SfN

In October 2019, I traveled to Chicago to present ongoing efforts on the characterization of morphology, electrophysiology, and molecular identity across cortical layers at Northwestern University and Neuroscience 2019.


Details

Let's get in touch!

If you think we should work together, discuss about research, or have an informal conversation, just drop me a line.