Dominic Kafka

Publication Activity (10 Years)

Years Active: 2019-2023
Publications (10 Years): 9

Top Topics

Loss Minimization

Rendezvous Search

Gradient Projection

Top Venues

J. Glob. Optim.

Publications

Younghwan Chae, Daniel N. Wilke, Dominic Kafka
Gradient-only surrogate to resolve learning rates for robust and consistent training of deep neural networks. Appl. Intell. 53 (11) (2023)
Dominic Kafka, Daniel N. Wilke
Resolving learning rates adaptively by locating stochastic non-negative associated gradient projection points using line searches. J. Glob. Optim. 79 (1) (2021)
Younghwan Chae, Daniel N. Wilke, Dominic Kafka
GOALS: Gradient-Only Approximations for Line Searches Towards Robust and Consistent Training of Deep Neural Networks. CoRR (2021)
Dominic Kafka, Daniel N. Wilke
An empirical study into finding optima in stochastic optimization of neural networks. Inf. Sci. 560 (2021)
Dominic Kafka, Daniel Nicolas Wilke
Gradient-only line searches to automatically determine learning rates for a variety of stochastic training algorithms. CoRR (2020)
Dominic Kafka, Daniel N. Wilke
Investigating the interaction between gradient-only line searches and different activation functions. CoRR (2020)
Dominic Kafka, Daniel N. Wilke
Resolving learning rates adaptively by locating Stochastic Non-Negative Associated Gradient Projection Points using line searches. CoRR (2020)
Dominic Kafka, Daniel N. Wilke
Visual interpretation of the robustness of Non-Negative Associative Gradient Projection Points over function minimizers in mini-batch sampled loss functions. CoRR (2019)
Dominic Kafka, Daniel N. Wilke
Gradient-only line searches: An Alternative to Probabilistic Line Searches. CoRR (2019)