Lukas Balles
7 papers · 2018–2025 · 4 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 ↓
🌍
Conference Polyglot · Bronze
🏃
Academic Marathon · Sprint
🌉
Interdisciplinary Bridge · Bronze
🗺️
Taxonomy Completionist
(17)
Venues
Top co-authors
Research topics
Machine Learning
(7)
Deep Learning
(5)
Optimization & Theory · Machine Learning
(4)
Optimization & Theory · Deep Learning
(2)
Artificial Intelligence
(1)
Computer Vision
(1)
Mathematics & Optimization
(1)
Natural Language Processing
(1)
Analysis
(1)
Application Areas
(1)
Architectures
(1)
Core AI
(1)
Core Methods
(1)
Generation
(1)
Learning Types
(1)
Keywords
neural network optimization
(2)
stochastic gradient descent
(2)
stochastic gradient
(1)
hyperparameter optimization
(1)
neural architecture search
(1)
deep learning
(1)
natural gradient
(1)
fisher information
(1)
gradient descent
(1)
resource allocation
(1)
natural gradient descent
(1)
motion segmentation
(1)
fisher information matrix
(1)
optical flow
(1)
early stopping
(1)
second-order optimization
(1)
adam optimizer
(1)
language model
(1)
second-order method
(1)
gradient variance
(1)
Papers
Hierarchical Autoregressive Transformers: Combining Byte- and Word-Level Processing for Robust, Adaptable Language Models
ICLR 2025
u-μP: The Unit-Scaled Maximal Update Parametrization
ICLR 2025
PASHA: Efficient HPO and NAS with Progressive Resource Allocation
ICLR 2023
Competitive Collaboration: Joint Unsupervised Learning of Depth, Camera Motion, Optical Flow and Motion Segmentation
CVPR 2019
DeepOBS: A Deep Learning Optimizer Benchmark Suite
ICLR 2019
Limitations of the empirical Fisher approximation for natural gradient descent
NeurIPS 2019
Dissecting Adam: The Sign, Magnitude and Variance of Stochastic Gradients
ICML 2018