Prediction-Oriented Bayesian Active Learning
- URL: http://arxiv.org/abs/2304.08151v1
- Date: Mon, 17 Apr 2023 10:59:57 GMT
- Title: Prediction-Oriented Bayesian Active Learning
- Authors: Freddie Bickford Smith, Andreas Kirsch, Sebastian Farquhar, Yarin Gal,
Adam Foster, Tom Rainforth
- Abstract summary: Expected predictive information gain (EPIG) is an acquisition function that measures information gain in the space of predictions rather than parameters.
EPIG leads to stronger predictive performance compared with BALD across a range of datasets and models.
- Score: 51.426960808684655
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Information-theoretic approaches to active learning have traditionally
focused on maximising the information gathered about the model parameters, most
commonly by optimising the BALD score. We highlight that this can be suboptimal
from the perspective of predictive performance. For example, BALD lacks a
notion of an input distribution and so is prone to prioritise data of limited
relevance. To address this we propose the expected predictive information gain
(EPIG), an acquisition function that measures information gain in the space of
predictions rather than parameters. We find that using EPIG leads to stronger
predictive performance compared with BALD across a range of datasets and
models, and thus provides an appealing drop-in replacement.
Related papers
- Prediction-Powered Adaptive Shrinkage Estimation [0.9208007322096532]
Prediction-Powered Adaptive Shrinkage (PAS) is a method that bridges PPI with empirical Bayes shrinkage to improve the estimation of multiple means.
PAS adapts to the reliability of the ML predictions and outperforms traditional and modern baselines in large-scale applications.
arXiv Detail & Related papers (2025-02-20T00:24:05Z) - UNEM: UNrolled Generalized EM for Transductive Few-Shot Learning [35.62208317531141]
We advocate and introduce the unrolling paradigm, also referred to as "learning to optimize"
Our unrolling approach covers various statistical feature distributions and pre-training paradigms.
We report comprehensive experiments, which cover a breadth of fine-grained downstream image classification tasks.
arXiv Detail & Related papers (2024-12-21T19:01:57Z) - Optimizing Sequential Recommendation Models with Scaling Laws and Approximate Entropy [104.48511402784763]
Performance Law for SR models aims to theoretically investigate and model the relationship between model performance and data quality.
We propose Approximate Entropy (ApEn) to assess data quality, presenting a more nuanced approach compared to traditional data quantity metrics.
arXiv Detail & Related papers (2024-11-30T10:56:30Z) - Optimizing LLMs with Direct Preferences: A Data Efficiency Perspective [4.548047308860141]
This study investigates the impact of different type of preference data on model performance.
It aims to reduce their dependency on extensive amounts of preference data, which is expensive to collect.
arXiv Detail & Related papers (2024-10-22T00:11:41Z) - Self-Augmented Preference Optimization: Off-Policy Paradigms for Language Model Alignment [104.18002641195442]
We introduce Self-Augmented Preference Optimization (SAPO), an effective and scalable training paradigm that does not require existing paired data.
Building on the self-play concept, which autonomously generates negative responses, we further incorporate an off-policy learning pipeline to enhance data exploration and exploitation.
arXiv Detail & Related papers (2024-05-31T14:21:04Z) - Informed Spectral Normalized Gaussian Processes for Trajectory Prediction [0.0]
We propose a novel regularization-based continual learning method for SNGPs.
Our proposal builds upon well-established methods and requires no rehearsal memory or parameter expansion.
We apply our informed SNGP model to the trajectory prediction problem in autonomous driving by integrating prior drivability knowledge.
arXiv Detail & Related papers (2024-03-18T17:05:24Z) - ASPEST: Bridging the Gap Between Active Learning and Selective
Prediction [56.001808843574395]
Selective prediction aims to learn a reliable model that abstains from making predictions when uncertain.
Active learning aims to lower the overall labeling effort, and hence human dependence, by querying the most informative examples.
In this work, we introduce a new learning paradigm, active selective prediction, which aims to query more informative samples from the shifted target domain.
arXiv Detail & Related papers (2023-04-07T23:51:07Z) - Improved prediction rule ensembling through model-based data generation [0.0]
Prediction rule ensembles (PRE) provide interpretable prediction models with relatively high accuracy.
PRE obtain a large set of decision rules from a (boosted) decision tree ensemble, and achieves sparsitythrough application of Lasso-penalized regression.
This article examines the use of surrogate modelsto improve performance of PRE, wherein the Lasso regression is trained with the help of a massivedataset.
arXiv Detail & Related papers (2021-09-28T12:44:10Z) - Supervised PCA: A Multiobjective Approach [70.99924195791532]
Methods for supervised principal component analysis (SPCA)
We propose a new method for SPCA that addresses both of these objectives jointly.
Our approach accommodates arbitrary supervised learning losses and, through a statistical reformulation, provides a novel low-rank extension of generalized linear models.
arXiv Detail & Related papers (2020-11-10T18:46:58Z) - Learnable Bernoulli Dropout for Bayesian Deep Learning [53.79615543862426]
Learnable Bernoulli dropout (LBD) is a new model-agnostic dropout scheme that considers the dropout rates as parameters jointly optimized with other model parameters.
LBD leads to improved accuracy and uncertainty estimates in image classification and semantic segmentation.
arXiv Detail & Related papers (2020-02-12T18:57:14Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.