Stanislaw Jastrzebski
7 papers · 2018–2022 · 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 ↓+2 more ↓ Show less ↑
🌉 Interdisciplinary Bridge · Bronze 🌍 Conference Polyglot · Silver 🗺️ Taxonomy Completionist (13) 🐣 Hot Topic Early Bird 🧭 Keyword Pioneer
🔥
Unstoppable · Bronze
❓
The Questioner
(2)
Venues
Top co-authors
Research topics
Machine Learning
(6)
Optimization & Theory
(3)
Deep Learning
(2)
Core Methods
(2)
Learning Types · Machine Learning
(2)
Artificial Intelligence
(1)
Natural Language Processing
(1)
Application Areas
(1)
Architectures
(1)
Core AI
(1)
Learning Types · Deep Learning
(1)
Resources & Methods
(1)
Techniques
(1)
Keywords
stochastic gradient descent
(2)
learning rate
(2)
transfer learning
(1)
text classification
(1)
knowledge mining
(1)
batch normalization
(1)
commonsense knowledge
(1)
multimodal learning
(1)
neural network optimization
(1)
loss landscape
(1)
novelty detection
(1)
fisher information matrix
(1)
high-dimensional optimization
(1)
knowledge base completion
(1)
deep neural network
(1)
implicit regularization
(1)
modality imbalance
(1)
mode connectivity
(1)
parameter efficiency
(1)
gradient variance
(1)
Papers
Characterizing and Overcoming the Greedy Nature of Learning in Multi-modal Deep Neural Networks
ICML 2022
Catastrophic Fisher Explosion: Early Phase Fisher Matrix Impacts Generalization
ICML 2021
Can Wikipedia Categories Improve Masked Language Model Pretraining?
ACL 2020
The Break-Even Point on Optimization Trajectories of Deep Neural Networks
ICLR 2020
Large Scale Structure of Neural Network Loss Landscapes
NeurIPS 2019
Parameter-Efficient Transfer Learning for NLP
ICML 2019
Commonsense mining as knowledge base completion? A study on the impact of novelty
NAACL 2018