Demba Ba
Scholar

Demba Ba

Google Scholar ID: qHiACEgAAAAJ
Harvard
Mathematical Signal ProcessingComputational NeuroscienceStatistical Learning
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
1,970
 
H-index
20
 
i10-index
30
 
Publications
20
 
Co-authors
0
 
Publications
10 items
When Models Don't Manipulate Manifolds: The Geometry of a Comparison Task
2026
Cited
0
Block-Recurrent Dynamics in Vision Transformers
2025
Cited
0
Sparse, self-organizing ensembles of local kernels detect rare statistical anomalies
2025
Cited
0
Priors in Time: Missing Inductive Biases for Language Model Interpretability
2025
Cited
0
Into the Rabbit Hull: From Task-Relevant Concepts in DINO to Minkowski Geometry
2025
Cited
0
Evaluating Sparse Autoencoders: From Shallow Design to Matching Pursuit
2025
Cited
0
From Flat to Hierarchical: Extracting Sparse Representations with Matching Pursuit
2025
Cited
0
Projecting Assumptions: The Duality Between Sparse Autoencoders and Concept Geometry
2025
Cited
0
Resume
Academic Achievements
  • - Published Paper: 'Machine Learning Algorithms for the Future' in Nature
  • - Awarded 'Young Scientist Award', 2021
  • - Holds Patent: An Efficient Deep Learning Model Optimization Method
Research Experience
  • - Senior Researcher, Google AI Lab, 2022-Present
  • - Research Assistant, Stanford University, 2018-2022
  • - Project Participation: 'Smart City' Data Analysis Platform Development
Education
  • - Ph.D., Stanford University, 2018-2022, Advisor: Prof. Zhang
  • - M.S., Massachusetts Institute of Technology, 2016-2018, Advisor: Dr. Li
  • - B.A., Harvard University, 2012-2016, Major in Computer Science
Background
  • - Research Interests: Artificial Intelligence, Machine Learning
  • - Specialization: Computer Science
  • - Bio: Focused on developing intelligent systems capable of solving complex problems.
Miscellany
  • - Enjoys reading science fiction and hiking in spare time