Picture for Balázs Kulcsár

Balázs Kulcsár

Less Is More -- On the Importance of Sparsification for Transformers and Graph Neural Networks for TSP

Mar 25, 2024
Figure 1 for Less Is More -- On the Importance of Sparsification for Transformers and Graph Neural Networks for TSP
Figure 2 for Less Is More -- On the Importance of Sparsification for Transformers and Graph Neural Networks for TSP
Figure 3 for Less Is More -- On the Importance of Sparsification for Transformers and Graph Neural Networks for TSP
Figure 4 for Less Is More -- On the Importance of Sparsification for Transformers and Graph Neural Networks for TSP
Viaarxiv icon

Critical Zones for Comfortable Collision Avoidance with a Leading Vehicle

Mar 26, 2023
Figure 1 for Critical Zones for Comfortable Collision Avoidance with a Leading Vehicle
Figure 2 for Critical Zones for Comfortable Collision Avoidance with a Leading Vehicle
Figure 3 for Critical Zones for Comfortable Collision Avoidance with a Leading Vehicle
Figure 4 for Critical Zones for Comfortable Collision Avoidance with a Leading Vehicle
Viaarxiv icon

Controlled Descent Training

Mar 16, 2023
Figure 1 for Controlled Descent Training
Figure 2 for Controlled Descent Training
Figure 3 for Controlled Descent Training
Figure 4 for Controlled Descent Training
Viaarxiv icon

Deep Q-learning: a robust control approach

Jan 21, 2022
Figure 1 for Deep Q-learning: a robust control approach
Figure 2 for Deep Q-learning: a robust control approach
Figure 3 for Deep Q-learning: a robust control approach
Figure 4 for Deep Q-learning: a robust control approach
Viaarxiv icon

Short-term traffic prediction using physics-aware neural networks

Sep 21, 2021
Figure 1 for Short-term traffic prediction using physics-aware neural networks
Figure 2 for Short-term traffic prediction using physics-aware neural networks
Figure 3 for Short-term traffic prediction using physics-aware neural networks
Figure 4 for Short-term traffic prediction using physics-aware neural networks
Viaarxiv icon

Constrained Policy Gradient Method for Safe and Fast Reinforcement Learning: a Neural Tangent Kernel Based Approach

Jul 19, 2021
Figure 1 for Constrained Policy Gradient Method for Safe and Fast Reinforcement Learning: a Neural Tangent Kernel Based Approach
Figure 2 for Constrained Policy Gradient Method for Safe and Fast Reinforcement Learning: a Neural Tangent Kernel Based Approach
Figure 3 for Constrained Policy Gradient Method for Safe and Fast Reinforcement Learning: a Neural Tangent Kernel Based Approach
Figure 4 for Constrained Policy Gradient Method for Safe and Fast Reinforcement Learning: a Neural Tangent Kernel Based Approach
Viaarxiv icon

Parameter and density estimation from real-world traffic data: A kinetic compartmental approach

Jan 27, 2021
Figure 1 for Parameter and density estimation from real-world traffic data: A kinetic compartmental approach
Figure 2 for Parameter and density estimation from real-world traffic data: A kinetic compartmental approach
Figure 3 for Parameter and density estimation from real-world traffic data: A kinetic compartmental approach
Figure 4 for Parameter and density estimation from real-world traffic data: A kinetic compartmental approach
Viaarxiv icon