Ai alignment: A comprehensive survey

J Ji, T Qiu, B Chen, B Zhang, H Lou, K Wang… - arxiv preprint arxiv …, 2023‏ - arxiv.org
AI alignment aims to make AI systems behave in line with human intentions and values. As
AI systems grow more capable, so do risks from misalignment. To provide a comprehensive …

Reimagining multi-criterion decision making by data-driven methods based on machine learning: A literature review

H Liao, Y He, X Wu, Z Wu, R Bausys - Information Fusion, 2023‏ - Elsevier
Multi-criterion decision making (MCDM) methods can derive alternative rankings as
solutions to decision-making problems based on survey or historical data about the …

Preference learning algorithms do not learn preference rankings

A Chen, S Malladi, L Zhang, X Chen… - Advances in …, 2025‏ - proceedings.neurips.cc
Preference learning algorithms (eg, RLHF and DPO) are frequently used to steer LLMs to
produce generations that are more preferred by humans, but our understanding of their …

Learning with differentiable pertubed optimizers

Q Berthet, M Blondel, O Teboul… - Advances in neural …, 2020‏ - proceedings.neurips.cc
Abstract Machine learning pipelines often rely on optimizers procedures to make discrete
decisions (eg, sorting, picking closest neighbors, or shortest paths). Although these discrete …

Fast differentiable sorting and ranking

M Blondel, O Teboul, Q Berthet… - … on Machine Learning, 2020‏ - proceedings.mlr.press
The sorting operation is one of the most commonly used building blocks in computer
programming. In machine learning, it is often used for robust statistics. However, seen as a …

A survey on multi-label feature selection from perspectives of label fusion

W Qian, J Huang, F Xu, W Shu, W Ding - Information Fusion, 2023‏ - Elsevier
With the rapid advancement of big data technology, high-dimensional datasets comprising
multi-label data have become prevalent in various fields. However, these datasets often …

[ספר][B] Active preference-based learning of reward functions

D Sadigh, A Dragan, S Sastry, S Seshia - 2017‏ - escholarship.org
Our goal is to efficiently learn reward functions encoding a human's preferences for how a
dynamical system should act. There are two challenges with this. First, in many problems it is …

Multilevel language and vision integration for text-to-clip retrieval

H Xu, K He, BA Plummer, L Sigal, S Sclaroff… - Proceedings of the …, 2019‏ - ojs.aaai.org
We address the problem of text-based activity retrieval in video. Given a sentence describing
an activity, our task is to retrieve matching clips from an untrimmed video. To capture the …

[PDF][PDF] Do we need hundreds of classifiers to solve real world classification problems?

M Fernández-Delgado, E Cernadas, S Barro… - The journal of machine …, 2014‏ - jmlr.org
We evaluate 179 classifiers arising from 17 families (discriminant analysis, Bayesian, neural
networks, support vector machines, decision trees, rule-based classifiers, boosting, bagging …

A survey of preference-based reinforcement learning methods

C Wirth, R Akrour, G Neumann, J Fürnkranz - Journal of Machine Learning …, 2017‏ - jmlr.org
Reinforcement learning (RL) techniques optimize the accumulated long-term reward of a
suitably chosen reward function. However, designing such a reward function often requires …