LOGICNN — PROGRAM
Days: Saturday, 25 July 2026
Saturday, 25 July 2026
09:00-10:10
Invited Talk 1
LOGICNN
Location:
C4.08
| 09:00-10:00 |
Logic and the Power of Recurrent Graph Neural Networks (abstract) 60 min
1 RWTH Aachen University, Germany
|
10:10-10:30
Coffee Break
LOGICNN
Location:
C4.08
10:30-11:10
Session 1A: Expressiveness
LOGICNN
Location:
C4.08
| 10:30-10:50 |
Expressive Power of Graph Transformers via Logic (abstract) 20 min
1 Tampere University
2 University of Leipzig
|
| 10:50-11:10 |
Towards Understanding the Expressive Power of GNNs with Global Readout (Extended Summary) (abstract) 20 min
1 Leipzig University
|
11:20-12:00
Session 1B: Expressiveness
LOGICNN
Location:
C4.08
| 11:20-11:40 |
An Algebraic Characterization of Local Weisfeiler–Leman (abstract) 20 min
1 Simon Fraser University
|
| 11:40-12:00 |
Words and Temporal Graphs: Comparing the Expressive Power of State Space Models and Recurrent Neural Networks (Extended Abstract) (abstract) 20 min
1 University of Kassel
|
12:00-13:30
Lunch
LOGICNN
Location:
C4.08
13:30-14:30
Invited Talk 2
LOGICNN
Location:
C4.08
| 13:30-14:30 |
Verification of DNNs with Marabou (abstract) 60 min
1 Hebrew University of Jerusalem, Israël
|
14:40-15:20
Session 2: Verification
LOGICNN
Location:
C4.08
| 14:40-15:00 |
Towards Continuous Constraint Programming for Sound Neural Network Verification (abstract) 20 min
1 University of Luxembourg
|
| 15:00-15:20 |
IsaGrad: Verified Automatic Differentiation over Computational Graphs in Imperative HOL (abstract) 20 min
1 University of Edinburgh
|
15:20-15:50
Coffee Break
LOGICNN
Location:
C4.08
15:50-16:50
Session 3: From Neural Networks to Logic
LOGICNN
Location:
C4.08
| 15:50-16:10 |
Neural networks as fuzzy logic formulas (abstract) 20 min
1 Tampere University
|
| 16:10-16:30 |
Neural Networks into Łukasiewicz Logic, with Applications to Formal Verification (abstract) 20 min
1 Federal University of ABC
2 University of São Paulo
|
| 16:30-16:50 |
From MLPs to Logic: An End-to-End Neuro-Symbolic Compilation Pipeline for Explainable Safety-Critical Medical AI (abstract) 20 min
1 University of Sussex
2 London South Bank University
|