FCAI Community Event and NeurIPS 2023

One hundred members of the FCAI community gathered at Aalto Design Factory on December 4 to share recent work, discuss new Teams and celebrate the submissions that are heading to NeurIPS — links to those are below. Here are a few photos from the event.

Below are links to the papers from Finland that will be presented at NeurIPS, 10-16 December 2023. NeurIPS is a major forum for leading research in machine learning and computational sciences.

• Beyond Uniform Sampling: Offline Reinforcement Learning with Imbalanced Datasets

• Characteristic Circuits

• Compositional Sculpting of Iterative Generative Processes

• Conditional Mutual Information for Disentangled Representations in Reinforcement Learning

• Continuous-Time Functional Diffusion Processes

• Generalized test utilities for long-tail performance in extreme multi-label classification

• Globally injective and bijective neural operators

• Going beyond persistent homology using persistent homology

• Hybrid Search for Efficient Planning with Completeness Guarantees

• Improving the Privacy and Practicality of Objective Perturbation for Differentially Private Linear Learners

• Learning Robust Statistics for Simulation-based Inference under Model Misspecification

• Learning Space-Time Continuous Latent Neural PDEs from Partially Observed States

• Practical Differentially Private Hyperparameter Tuning with Subsampling

• Practical Equivariances via Relational Conditional Neural Processes

• Provable benefits of annealing for estimating normalizing constants: Importance Sampling, Noise-Contrastive Estimation, and beyond

• Temporal Causal Mediation through a Point Process: Direct and Indirect Effects of Healthcare Interventions

• The Cambridge Law Corpus: A Corpus for Legal AI Research

• Thin and deep Gaussian processes

• LART: Neural Correspondence Learning with Latent Regularization Transformer for 3D Motion Transfer

• Scaling Data-Constrained Language Models

• STARSS23: An Audio-Visual Dataset of Spatial Recordings of Real Scenes with Spatiotemporal Annotations of Sound Events