Pierre Baldi
Distinguished Professor of Computer Science and Director of the Artificial Intelligence in Science Institute UC Irvine
- Irvine CA
Pierre Baldi works at the intersection of biology and computer science with an expertise in artificial intelligence and deep learning.
Media
Social
Biography
Dr. Baldi's group works at the intersection of biological and computer sciences, using probabilistic/machine learning techniques to address biological problems and mine large data sets produced by massive data acquisition technologies, such as genome sequencing, high-throughput drug screening, and DNA microarrays. Current projects include the prediction of protein secondary and tertiary structure, the study of DNA structure in relation to several biological processes (protein binding, gene regulation, triplet repeat expansion diseases), and the analysis of gene expression data. Dr. Baldi's group is building a suite of genomics and proteomics programs for the prediction of protein structure and function and the analysis of microarray data.
Areas of Expertise
Accomplishments
Dennis Gabor Award of the International Neural Network Society
2023
Fellow, Asia-Pacific Artificial Intelligence Association (AAIA)
2022
Prominent Artificial Intelligence Journal Paper Award
2019
Named Top 100 AI Leaders in Drug Discovery and Advanced Healthcare, Deep Knowledge Analytics
2019
Education
California Institute of Technology
Ph.D.
Mathematics
ENSTA Paris
M.S.
Computer Science and Engineering
1983
University of Paris VII
D.E.A.
Mathematics
1981
University of Paris VII
M.S.
Mathematics
1980
University of Paris X
M.S.
Psychology
1980
Affiliations
- American Association for the Advancement of Science (AAAS)
- Association for the Advancement of Artificial Intelligence (AAAI)
- Institute of Electrical and Electronic Engineers (IEEE)
- American Chemical Society (ACS)
- Association for Computing Machinery (ACM)
- International Society for Computational Biology (ISCB)
Media Appearances
Pierre Baldi (PhD '86), Computer Scientist and Explorer of the Natural-Artifical Intelligence Interface
Caltech | Heritage Project online
2023-12-21
While these are discrete fields, some of the most important work is happening at the interface, and the research benefits both in natural and in artificial intelligence has been astounding. At the center of these developments is Caltech alumnus and UC Irvine Distinguished Professor Pierre Baldi, whose educational trajectory and research achievements have had - and will have for decades to come - a major impact on what intelligence means and what it can accomplish.
A call to hit the ‘pause’ button on AI experiments
University of California online
2023-04-07
“I signed the letter too on the first day,” says Pierre Baldi, distinguished professor of computer science in UC Irvine’s Donald Bren School of Information and Computer Sciences (ICS). “I have some doubts as to whether a real pause can be implemented worldwide, but even if it cannot, the letter is useful for raising awareness about the issues.”
Researchers' Deep Learning Algorithm Solves Rubik's Cube Faster Than Any Human
Lab Manager online
2019-07-15
The fastest people need about 50 moves to solve a Rubik's Cube. "Our AI takes about 20 moves, most of the time solving it in the minimum number of steps," says the study's senior author, Pierre Baldi, UCI distinguished professor of computer science. "Right there, you can see the strategy is different, so my best guess is that the AI's form of reasoning is completely different from a human's."
Event Appearances
Workshop on Deep Learning: Theory, Algorithms, and Implications
2025 Rome, Italy
Workshop on Deep Learning: Theory, Algorithms, and Implications
2024 Tokyo, Japan
Workshop on Deep Learning: Theory, Algorithms, and Applications
2023 Trento, Italy
Research Grants
R01GM123558
NIH
6/1/23-5/31/27
RP1 EB029751
NIH
5/1/20-1/31/24
HL128801
NIH
5/1/21-4/30/22
Articles
Domain Knowledge Inclusive Monotonic Neural Network Guides Patient-Specific Induction of General Anesthesia Dosing
Kathryn Sarullo, Muntaha Samad, Samir Kendale, Pierre Baldi, S Joshua SwamidassKathryn Sarullo, Muntaha Samad, Samir Kendale, Pierre Baldi, S Joshua Swamidass
2025
Postinduction hypotension is a well-known risk factor for adverse postoperative outcomes. Anesthesiologists estimate anesthetic dosages based on a patient’s chart and domain knowledge. Machine learning is increasingly applied in predicting postinduction hypotension, with neural networks providing a robust and accurate approach to model complex relationships. This study aims to use machine learning to suggest anesthetic doses, both generalized to an average patient population and personalized for specific patients, incorporating domain knowledge into the modeling process.
Particle hit clustering and identification using point set transformers in liquid argon time projection chambers
Journal of InstrumentationEdgar E Robles, Alejandro Yankelevich, Wenjie Wu, Jianming Bian, Pierre Baldi
2025
Liquid argon time projection chambers are often used in neutrino physics and dark-matter searches because of their high spatial resolution. The images generated by these detectors are extremely sparse, as the energy values detected by most of the detector are equal to 0, meaning that despite their high resolution, most of the detector is unused in a particular interaction. Instead of representing all of the empty detections, the interaction is usually stored as a sparse matrix, a list of detection locations paired with their energy values. Traditional machine learning methods that have been applied to particle reconstruction such as convolutional neural networks (CNNs), however, cannot operate over data stored in this way and therefore must have the matrix fully instantiated as a dense matrix.
Cilia in the brain display region-dependent oscillations of length and orientation
PLoS biologyRoudabeh Vakil Monfared, Sherif Abdelkarim, Pieter Derdeyn, Kiki Chen, Hanting Wu, Kenneth Leong, Tiffany Chang, Justine Lee, Sara Versales, Surya M Nauli, Kevin Beier, Pierre Baldi, Amal Alachkar
2025
In this study, we conducted high-throughput spatiotemporal analysis of primary cilia length and orientation across 22 mouse brain regions. We developed automated image analysis algorithms, which enabled us to examine over 10 million individual cilia, generating the largest spatiotemporal atlas of cilia. We found that cilia length and orientation display substantial variations across different brain regions and exhibit fluctuations over a 24-h period, with region-specific peaks during light-dark phases. Our analysis revealed unique orientation patterns of cilia, suggesting that cilia orientation within the brain is not random but follows specific patterns. Using BioCycle, we identified rhythmic fluctuations in cilia length across five brain regions: the nucleus accumbens core, somatosensory cortex, and the dorsomedial, ventromedial, and arcuate hypothalamic nuclei.
Evaluating the Intelligence of large language models: A comparative study using verbal and visual IQ tests
Computers in Human Behavior: Artificial HumansSherif Abdelkarim, David Lu, Dora-Luz Flores, Susanne Jaeggi, Pierre Baldi
2025
Large language models (LLMs) excel on many specialised benchmarks, yet their general-reasoning ability remains opaque. We therefore test 18 models—including GPT-4, Claude 3 and Gemini Pro—on a 14-section IQ suite spanning verbal, numerical and visual puzzles and add a” multi-agent reflection” variant in which one model answers while others critique and revise. Results replicate known patterns: a strong bias towards verbal vs numerical reasoning (GPT-4: 79% vs 53% accuracy), a pronounced modality gap (text-IQ≈ 125 vs visual-IQ≈ 103), and persistent failure on abstract arithmetic (≤ 20% on missing-number tasks). Scaling lifts mean IQ from 89 (tiny models) to 131 (large models), but gains are non-uniform, and reflection yields only modest extra points for frontier systems.
Succinate Modulation as a Novel Mechanism Underlying the Effects of Intermittent Fasting on Brain Function and Metabolism in Diet-Induced Obesity
bioRxivAndrea Tognozzi, Fabrizia Carli, Sherif Abdelkarim, Sara Cornuti, Francesca Damiani, Maria Grazia Giuliano, Alice Miniati, Martina Nasisi, Lia De Benedictis, Kousha Changizi Ashtiani, Gaia Scabia, Margherita Maffei, Pierre Baldi, Amalia Gastaldelli, Paola Tognini
2025
Obesity significantly impacts the central nervous system (CNS), increasing risks of neuropsychiatric disorders and dementia. Intermittent fasting (IF) shows promise for improving peripheral and CNS health, but its mechanisms are unclear. Using a diet-induced obesity mouse model (10 weeks high fat diet (HFD), then 4 weeks intervention), we compared HFD, HFD-IF, ad libitum control chow (CC), and CC-IF groups. Switching to CC or IF reduced body weight, fat mass, and improved glucose tolerance. Notably, CC-IF uniquely enhanced exploration and reduced anxiety-like behavior. Transcriptomics revealed HFD-induced hippocampal neuroinflammation, while metabolomics identified a specific succinate signature in CC-IF mice: plasma concentration decreased while liver and brown adipose tissue (BAT) levels increased. Succinate supplementation mimicked CC-IF metabolic and behavioral benefits and reduced hippocampal inflammation.
