ECAI 2020

212 papers
The 2020 edition of the European Conference on Artificial Intelligence (ECAI) featured 292 papers, with a strong emphasis on neuro-symbolic integration and the formal verification of neural networks. Research themes prominently included the development of architectures that decouple perception from reasoning and the scaling of safety certificates for uncertain dynamical systems. Significant attention was also directed toward interpretability, specifically bridging the gap between algorithmic outputs and human understanding while addressing the performance costs associated with interpretable models. Technical contributions spanned a variety of domains, including adversarial domain adaptation, tensor-based computation, and link prediction in multiplex networks. Applied research was also represented, such as the use of mobile phone records to estimate malaria parasite mobility in Mozambique. Prolific authors contributing to the proceedings included Meng Wang, Yang Wang, Zhao Zhang, and Francesco Belardinelli. These works collectively highlighted a trend toward making machine learning systems more verifiable, explainable, and sample-efficient through the combination of neural and symbolic methodologies.

Top Research Directions

Research directions most represented by ECAI 2020 papers.

Generalizing Machine Learning Frameworks Across Network Architectures 2
Robust Feature Alignment for Unsupervised Domain Adaptation 2
Decoupling Search from Storage in Multi-Objective Optimization 1
Learned Clause Retention in Incremental Constraint Solving 1
Hierarchical Feature Specialization in Deep Networks 1
Rational Verification of Multi Agent Systems 1
Diagnostic Trace Generation in Formal Verification 1
Enforcing Bidirectional Consistency in Latent Spaces 1
Object Counting via Spatial Density Estimation 1
Pinpointing Minimal Conflicts in Overconstrained Systems 1
Unsupervised Discovery of Latent Behavioral Personas 1
Relational Invariance for Robust Domain Adaptation 1
Data Efficiency Through Neuro Symbolic Integration 1
Solving Word Equations with Length Constraints 1
Tracking Dialogue Progress with Discrete State Transitions 1

Top Authors

AuthorPapers
Jose Hernandez-Orallo4
Fernando Martínez-Plumed3
Francesco Belardinelli3
Frank Neumann3
Meng Wang3
Nuria M Oliver3
Thomas Eiter3
Zhao Zhang3
Alessandro Sperduti2
Aneta Neumann2
Claudia Plant2
Deyi Xiong2
Fausto Giunchiglia2
Giuseppe Marra2
Hisao Ishibuchi2
Jingyu Wang2
Jiuyong Li2
Jong-Seok Lee2
Jörg Rothe2
Jun Zhou2

Impactful Papers

Papers from ECAI 2020 that shaped research clusters.

Foundational

Influential

Papers

Title
OpenSMax: Unknown Domain Generation Algorithm Detection
Outside the Box: Abstraction-Based Monitoring of Neural Networks
Parameterized Complexity of Manipulating Sequential Allocation
Partial Label Learning via Generative Adversarial Nets
Personalized Automated Machine Learning
Polynomial-Time Implicit Learnability in SMT
PushNet: Efficient and Adaptive Neural Message Passing
Randomized Kernel Multi-View Discriminant Analysis
Reasoning with a Bounded Number of Resources in ATL+
Reasoning with Justifiable Exceptions in Contextual Hierarchies
REBA-KRL: Refinement-Based Architecture for Knowledge Representation, Explainable Reasoning and Interactive Learning in Robotics
Regularized Cycle Consistent Generative Adversarial Network for Anomaly Detection
Relational Neural Machines
Representing and Reducing Uncertainty for Enumerating the Belief Space to Improve Endgame Play in Skat
Retrospective and Prospective Mixture-of-Generators for Task-Oriented Dialogue Response Generation
Robust Multi-Instance Learning with Stable Instances
Robust Submodular Minimization with Applications to Cooperative Modeling
Saliency Detection with Deformable Convolution and Feature Attention
Search for Better Students to Learn Distilled Knowledge
Secure Social Recommendation Based on Secret Sharing
Self-Attention-Based Fully-Inception Networks for Continuous Sign Language Recognition
Semantic Point Completion Network for 3D Semantic Scene Completion
Semantics-Reinforced Networks for Question Generation
SETNet: A Novel Semi-Supervised Approach for Semantic Parsing
SI-AGAN: Spatial Interpolation with Attentional Generative Adversarial Networks for Environment Monitoring
Simplifying Graph Attention Networks with Source-Target Separation
Speeding up Word Mover's Distance and Its Variants via Properties of Distances Between Embeddings
Step-Wise Explanations of Constraint Satisfaction Problems
ST-MFM: A Spatiotemporal Multi-Modal Fusion Model for Urban Anomalies Prediction
Stochastic Local Search and Machine Learning: From Theory to Applications and Vice Versa
Strategic Monitor Placement Against Malicious Flows
STRiKE: Rule-Driven Relational Learning Using Stratified k-Entailment
String Constraint Solving: Past, Present and Future
Strong Statistical Correlation Revealed by Quantum Entanglement for Supervised Learning
Structure Matters: Towards Generating Transferable Adversarial Images
Tackling Imitative Attacker Deception in Repeated Bayesian Stackelberg Security Games
Task-Aware Performance Prediction for Efficient Architecture Search
The Automated Inspection of Opaque Liquid Vaccines
The Challenge of Optimal Paths in Graphs with Item Sets
The Last Voting Rule Is Home: Complexity of Control by Partition of Candidates or Voters in Maximin Elections
The Quality of Content Publishing in the Digital Era
TIMELY: Improving Labeling Consistency in Medical Imaging for Cell Type Classification
Topological Bayesian Optimization with Persistence Diagrams
Toward Metrics for Differentiating Out-of-Distribution Sets
Towards Efficient Discrete Integration via Adaptive Quantile Queries
Towards Meta-Reasoning for Ontologies: A Roadmap
Tracking AI: The Capability Is (Not) Near
TransSketchNet: Attention-Based Sketch Recognition Using Transformers
TREPAN Reloaded: A Knowledge-Driven Approach to Explaining Black-Box Models
Understanding and Training Deep Diagonal Circulant Neural Networks

« prev Page 4/5 next »