An official website of the United States government
Many sequential decision making tasks can be viewed as combinatorial optimiza- tion problems over a large number of actions. When the cost of evaluating an ac- tion is high, even a greedy algorithm, which iteratively picks the best action given the history, is prohibitive to run. In this paper, we aim to learn a greedy heuris- tic for sequentially selecting actions as a surrogate for invoking the expensive oracle when evaluating an action. In particular, we focus on a class of combinato- rial problems that can be solved via submodular maximization (either directly on the objective function or via submodular surrogates). We introduce a data-driven optimization framework based on the submodular-norm loss, a novel loss func- tion that encourages the resulting objective to exhibit diminishing returns. Our framework outputs a surrogate objective that is efficient to train, approximately submodular, and can be made permutation-invariant. The latter two properties al- low us to prove strong approximation guarantees for the learned greedy heuristic. Furthermore, our model is easily integrated with modern deep imitation learning pipelines for sequential prediction tasks. We demonstrate the performance of our algorithm on a variety of batched and sequential optimization tasks, including set cover, active learning, and data-driven protein engineering.
more »
« less
Learning to Rank for One-Round Active Learning
Active learning is a promising paradigm to reduce the labeling cost by strategically requesting labels to improve model performance. However, existing active learning methods often rely on expensive acquisition function to compute, extensive modeling retraining and multiple rounds of interaction with annotators. To address these limitations, we propose a novel approach for active learning, which aims to select batches of unlabeled instances through a learned surrogate model for data acquisition. A key challenge in this approach is developing an acquisition function that generalizes well, as the history of data, which forms part of the utility function’s input, grows over time. Our novel algorithmic contribution is a multi-task bilevel optimization framework that predicts the relative utility, measured by the validation accuracy, of different training sets, and ensures the learned acquisition function generalizes effectively. For cases where validation accuracy is expensive to evaluate, we introduce efficient interpolation-based surrogate models to estimate the utility function, reducing the evaluation cost. We demonstrate the performance of our approach through extensive experiments on standard active classification benchmarks.
more »
« less
- Award ID(s):
- 2313131
- PAR ID:
- 10528086
- Publisher / Repository:
- ICLR 2024 Workshop on Data-centric Machine Learning Research (DMLR): Harnessing Momentum for Science
- Date Published:
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Labeled data can be expensive to acquire in several application domains, including medical imaging, robotics, computer vision and wireless networks to list a few. To efficiently train machine learning models under such high labeling costs, active learning (AL) judiciously selects the most informative data instances to label on-the-fly. This active sampling process can benefit from a statistical function model, that is typically captured by a Gaussian process (GP) with well-documented merits especially in the regression task. While most GP-based AL approaches rely on a single kernel function, the present contribution advocates an ensemble of GP (EGP) models with weights adapted to the labeled data collected incrementally. Building on this novel EGP model, a suite of acquisition functions emerges based on the uncertainty and disagreement rules. An adaptively weighted ensemble of EGP-based acquisition functions is advocated to further robustify performance. Extensive tests on synthetic and real datasets in the regression task showcase the merits of the proposed EGP-based approaches with respect to the single GP-based AL alternatives.more » « less
-
Distributed multi-agent unmanned aerial systems (UAS) have the potential to be heavily utilized in environmental monitoring, especially in wetland monitoring. Deep active learning algorithms provide key tools to analyze the sensed images captured during monitoring and interpret them precisely. However, these algorithms demand significant computational resources that limit their use with distributed UAS. In this paper, we propose a novel algorithm for consensus-enabled active learning that drastically reduces the computational demand while increasing the overall model accuracy. Once each of the UAS obtains a labeled subset of images through active learning, we update the weights of the model for three epochs only on the new images to reduce the computational cost, allowing for an increased operational time. The group of UAS communicates the model weights instead of the raw data and then leverages consensus to agree on updated weights. The consensus step mitigates the impact on weights caused by the updates and generalizes the knowledge of each individual UAS to the whole system, which results in increased model accuracy. Our method achieved an average of 11.15% increase in accuracy over 25 acquisition iterations whilst utilizing only 16.8% of the processor time compared to the centralized method of active learning.more » « less
-
We develop novel methodology for active feature acquisition (AFA), the study of sequentially acquiring a dynamic subset of features that minimizes acquisition costs whilst still yielding accurate inference. The AFA framework can be useful in a myriad of domains, including health care applications where the cost of acquiring additional features for a patient (in terms of time, money, risk, etc.) can be weighed against the expected improvement to diagnostic performance. Previous approaches for AFA have employed either: deep learning RL techniques, which have difficulty training policies due to a complicated state and action space; deep learning surrogate generative models, which require modeling complicated multidimensional conditional distributions; or greedy policies, which cannot account for jointly informative feature acquisitions. We show that we can bypass many of these challenges with a novel, nonparametric oracle based approach, which we coin the acquisition conditioned oracle (ACO). Extensive experiments show the superiority of the ACO to state-of-the-art AFA methods when acquiring features for both predictions and general decision-making.more » « less
-
Optimizing a black-box function that is expensive to evaluate emerges in a gamut of machine learning and artifcial intelligence applications including drug discovery, policy optimization in robotics, and hyperparameter tuning of learning models to list a few. Bayesian optimization (BO) provides a principled framework to fnd the global optimum of such functions using a limited number of function evaluations. BO relies on a statistical surrogate model to actively select new query points, that is typically captured by a Gaussian process (GP). Unlike most existing approaches that hinge on a single GP surrogate model with a pre-selected kernel function that may confne the expressiveness of the sought function especially under the limited evaluation budget, the present work puts forth a weighted ensemble of GPs as a surrogate model. Building on the advocated Gaussian mixture (GM) posterior, the EGP framework adapts to the most ftted surrogate model as data arrive on-the-fy, offering a richer function space. For the acquisition of next evaluation points, the EGP-based posterior is coupled with an adaptive expected improvement (EI) criterion to balance exploration and exploitation of the search space. Numerical tests on a set of benchmark synthetic functions and two robotic tasks, demonstrate the impressive benefts of the proposed approach.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Workshop Report:
- The DOI is not currently available.
