↓Skip to main content

🏢 Aalto University

Preference Learning of Latent Decision Utilities with a Human-like Model of Preferential Choice
·2258 words·11 mins· loading · loading
Machine Learning Reinforcement Learning 🏢 Aalto University
Human-like choice modeling revolutionizes preference learning! A new tractable model, CRCS, significantly improves utility inference from human data, outperforming existing methods.
Learning Structure-Aware Representations of Dependent Types
·1855 words·9 mins· loading · loading
AI Theory Representation Learning 🏢 Aalto University
This research pioneers the integration of machine learning with the dependently-typed programming language Agda, introducing a novel dataset and neural architecture for faithful program representation…
Improving robustness to corruptions with multiplicative weight perturbations
·1713 words·9 mins· loading · loading
Image Classification 🏢 Aalto University
Boost DNN robustness to corruptions without sacrificing clean image accuracy using Data Augmentation via Multiplicative Perturbations (DAMP)!
Generating Code World Models with Large Language Models Guided by Monte Carlo Tree Search
·4163 words·20 mins· loading · loading
AI Generated Machine Learning Reinforcement Learning 🏢 Aalto University
LLMs guided by Monte Carlo Tree Search generate precise, efficient Python code as world models for model-based reinforcement learning, significantly improving sample efficiency and inference speed.
Applying Guidance in a Limited Interval Improves Sample and Distribution Quality in Diffusion Models
·2501 words·12 mins· loading · loading
Computer Vision Image Generation 🏢 Aalto University
Boosting image generation: Applying guidance selectively during diffusion model sampling drastically enhances image quality and inference speed, achieving state-of-the-art results.
Amortized Bayesian Experimental Design for Decision-Making
·1746 words·9 mins· loading · loading
Machine Learning Active Learning 🏢 Aalto University
Amortized Decision-Aware BED prioritizes maximizing downstream decision utility by instantly proposing informative experimental designs and inferring decisions, facilitating accurate decision-making.
Algebraic Positional Encodings
·1392 words·7 mins· loading · loading
Machine Translation 🏢 Aalto University
Revolutionizing Transformers, Algebraic Positional Encodings (APE) offers a theory-first approach to positional encoding, outperforming state-of-the-art methods without hyperparameter tuning across va…