Michael Gastpar
12 papers · 2021–2025 · 5 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓+4 more ↓ Show less ↑
π Interdisciplinary Bridge Β· Silver πΊοΈ Taxonomy Completionist (26) π Conference Polyglot Β· Silver π Triple Crown π Century Club Β· Bronze
ποΈ
Keyword Collector
(66)
π₯
Unstoppable Β· Bronze
β‘
Prolific Year Β· Productive
π
Renaissance Researcher
(6)
Venues
Top co-authors
Research topics
Machine Learning
(11)
Optimization & Theory
(10)
Artificial Intelligence
(4)
Deep Learning
(4)
Architectures
(4)
Core AI
(4)
Application Areas
(3)
Mathematics & Optimization
(2)
Natural Language Processing
(2)
Generation
(2)
Optimization
(2)
Computer Science
(1)
Data Science & Analytics
(1)
Security & Privacy
(1)
Core Methods
(1)
Keywords
markov chain
(2)
large language model
(2)
information theory
(2)
in-context learning
(2)
local minimum
(2)
generalization error
(1)
attention mechanism
(1)
representation learning
(1)
feature representation
(1)
loss landscape
(1)
compression scheme
(1)
renyi divergence
(1)
parameter estimation
(1)
efficient inference
(1)
reward function
(1)
sparsity
(1)
total variation
(1)
rate distortion
(1)
local privacy
(1)
sample complexity
(1)
Papers
Attention with Markov: A Curious Case of Single-layer Transformers
ICLR 2025
Leveraging Sparsity for Sample-Efficient Preference Learning: A Theoretical Perspective
ICML 2025
Fantastic Generalization Measures are Nowhere to be Found
ICLR 2024
Fundamental Limits of Prompt Compression: A Rate-Distortion Framework for Black-Box Language Models
NeurIPS 2024
LASER: Linear Compression in Wireless Distributed Optimization
ICML 2024
Local to Global: Learning Dynamics and Effect of Initialization for Transformers
NeurIPS 2024
Lower Bounds on the Bayesian Risk via Information Measures
JMLR 2024
The Fundamental Limits of Least-Privilege Learning
ICML 2024
Transformers on Markov data: Constant depth suffices
NeurIPS 2024
Generalization Error Bounds for Noisy, Iterative Algorithms via Maximal Leakage
COLT 2023
A Johnson-Lindenstrauss Framework for Randomly Initialized CNNs
ICLR 2022
Locally Differentially-Private Randomized Response for Discrete Distribution Learning
JMLR 2021