Alert button
Picture for Daniel Justus

Daniel Justus

Alert button

BESS: Balanced Entity Sampling and Sharing for Large-Scale Knowledge Graph Completion

Nov 22, 2022
Alberto Cattaneo, Daniel Justus, Harry Mellor, Douglas Orr, Jerome Maloberti, Zhenying Liu, Thorin Farnsworth, Andrew Fitzgibbon, Blazej Banaszewski, Carlo Luschi

Figure 1 for BESS: Balanced Entity Sampling and Sharing for Large-Scale Knowledge Graph Completion
Figure 2 for BESS: Balanced Entity Sampling and Sharing for Large-Scale Knowledge Graph Completion
Figure 3 for BESS: Balanced Entity Sampling and Sharing for Large-Scale Knowledge Graph Completion
Figure 4 for BESS: Balanced Entity Sampling and Sharing for Large-Scale Knowledge Graph Completion
Viaarxiv icon

8-bit Numerical Formats for Deep Neural Networks

Jun 06, 2022
Badreddine Noune, Philip Jones, Daniel Justus, Dominic Masters, Carlo Luschi

Figure 1 for 8-bit Numerical Formats for Deep Neural Networks
Figure 2 for 8-bit Numerical Formats for Deep Neural Networks
Figure 3 for 8-bit Numerical Formats for Deep Neural Networks
Figure 4 for 8-bit Numerical Formats for Deep Neural Networks
Viaarxiv icon

Towards Structured Dynamic Sparse Pre-Training of BERT

Aug 13, 2021
Anastasia Dietrich, Frithjof Gressmann, Douglas Orr, Ivan Chelombiev, Daniel Justus, Carlo Luschi

Figure 1 for Towards Structured Dynamic Sparse Pre-Training of BERT
Figure 2 for Towards Structured Dynamic Sparse Pre-Training of BERT
Figure 3 for Towards Structured Dynamic Sparse Pre-Training of BERT
Figure 4 for Towards Structured Dynamic Sparse Pre-Training of BERT
Viaarxiv icon

GroupBERT: Enhanced Transformer Architecture with Efficient Grouped Structures

Jun 10, 2021
Ivan Chelombiev, Daniel Justus, Douglas Orr, Anastasia Dietrich, Frithjof Gressmann, Alexandros Koliousis, Carlo Luschi

Figure 1 for GroupBERT: Enhanced Transformer Architecture with Efficient Grouped Structures
Figure 2 for GroupBERT: Enhanced Transformer Architecture with Efficient Grouped Structures
Figure 3 for GroupBERT: Enhanced Transformer Architecture with Efficient Grouped Structures
Figure 4 for GroupBERT: Enhanced Transformer Architecture with Efficient Grouped Structures
Viaarxiv icon

Predicting the Computational Cost of Deep Learning Models

Nov 28, 2018
Daniel Justus, John Brennan, Stephen Bonner, Andrew Stephen McGough

Figure 1 for Predicting the Computational Cost of Deep Learning Models
Figure 2 for Predicting the Computational Cost of Deep Learning Models
Figure 3 for Predicting the Computational Cost of Deep Learning Models
Figure 4 for Predicting the Computational Cost of Deep Learning Models
Viaarxiv icon