检索结果-内蒙古大学图书馆

learning and fine-tuning a generic value-selection heuristic inside a constraint programming solver

CONSTRAINTS 2024年第3-4期29卷 234-260页

作者： Marty, Tom Boisvert, Leo Francois, Tristan Tessier, Pierre Gautier, Louis Rousseau, Louis-Martin Cappart, Quentin Polytech Montreal Montreal PQ Canada Ecole Polytech Palaiseau France Quebec Inst Learning Algorithms MILA Montreal PQ Canada

Constraint programming is known for being an efficient approach to solving combinatorial problems. Important design choices in a solver are the branching heuristics, designed to lead the search to the best solutions in a minimum amount of time. However, developing these heuristics is a time-consuming process that requires problem-specific expertise. This observation has motivated many efforts to use machine learning to automatically learn efficient heuristics without expert intervention. Although several generic variable-selection heuristics are available in the literature, the options for value-selection heuristics are more scarce. We propose to tackle this issue by introducing a generic learning procedure that can be used to obtain a value-selection heuristic inside a constraint programming solver. This has been achieved thanks to the combination of a deep Q-learning algorithm, a tailored reward signal, and a heterogeneous graph neural network. Experiments on graph coloring, maximum independent set, maximum cut, and minimum vertex cover problems show that this framework competes with the well-known impact-based and activity-based search heuristics and can find solutions close to optimality without requiring a large number of backtracks. Additionally, we observe that fine-tuning a model with a different problem class can accelerate the learning process.

关键词： Constraint programming Branching heuristics Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Behavioural equivalences for continuous-time Markov processes

引用

MATHEMATICAL STRUCTURES IN COMPUTER SCIENCE 2023年第4-5期33卷 222-258页

作者： Chen, Linan Clerc, Florence Panangaden, Prakash McGill Univ Dept Math & Stat Montreal PQ Canada Sch Comp Sci McGil lUnivers DEEL Quebec Montreal PQ Canada McGill Univ Montreal Inst Learning Algorithms MILA Sch Comp Sci Montreal PQ Canada

Bisimulation is a concept that captures behavioural equivalence of states in a variety of types of transition systems. It has been widely studied in a discrete-time setting. The core of this work is to generalise the discrete-time picture to continuous time by providing a notion of behavioural equivalence for continuous-time Markov processes. In Chen et al. [(2019). Electronic Notes in Theoretical Computer Science 347 45-63.], we proposed two equivalent definitions of bisimulation for continuous-time stochastic processes where the evolution is a flow through time: the first one as an equivalence relation and the second one as a cospan of morphisms. In Chen et al. [(2020). Electronic Notes in Theoretical Computer Science.], we developed the theory further: we introduced different concepts that correspond to different behavioural equivalences and compared them to bisimulation. In particular, we studied the relation between bisimulation and symmetry groups of the dynamics. We also provided a game interpretation for two of the behavioural equivalences. The present work unifies the cited conference presentations and gives detailed proofs.

关键词： Stochastic processes Markov processes continuous time bisimulation diffusion

来源：评论

学校读者我要写书评

暂无评论

Scaling Vision-based End-to-End Autonomous Driving with Multi-View Attention learning

Scaling Vision-based End-to-End Autonomous Driving with Mult...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Xiao, Yi Codevilla, Felipe Porres, Diego Lopez, Antonio M. Univ Autonoma Barcelona UAB Dept Comp Sci Comp Vis Ctr CVC Barcelona Spain Montreal Inst Learning Algorithms MILA Montreal PQ Canada

ISBN: (纸本)9781665491907

On end-to-end driving, human driving demonstrations are used to train perception-based driving models by imitation learning. This process is supervised on vehicle signals (e.g., steering angle, acceleration) but does not require extra costly supervision (human labeling of sensor data). As a representative of such vision-based end-to-end driving models, CILRS is commonly used as a baseline to compare with new driving models. So far, some latest models achieve better performance than CILRS by using expensive sensor suites and/or by using large amounts of human-labeled data for training. Given the difference in performance, one may think that it is not worth pursuing vision-based pure end-to-end driving. However, we argue that this approach still has great value and potential considering cost and maintenance. In this paper, we present CIL++, which improves on CILRS by both processing higher-resolution images using a human-inspired HFOV as an inductive bias and incorporating a proper attention mechanism. CIL++ achieves competitive performance compared to models which are more costly to develop. We propose to replace CILRS with CIL++ as a strong vision-based pure end-to-end driving baseline supervised by only vehicle signals and trained by conditional imitation learning.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Interpolation consistency training for semi-supervised learning

引用

NEURAL NETWORKS 2022年 145卷 90-106页

作者： Verma, Vikas Kawaguchi, Kenji Lamb, Alex Kannala, Juho Solin, Arno Bengio, Yoshua Lopez-Paz, David Montreal Inst Learning Algorithms MILA Montreal PQ Canada Aalto Univ Espoo Finland Harvard Univ Cambridge MA 02138 USA Facebook AI Res Paris France

We introduce Interpolation Consistency Training (ICT), a simple and computation efficient algorithm for training Deep Neural Networks in the semi-supervised learning paradigm. ICT encourages the prediction at an interpolation of unlabeled points to be consistent with the interpolation of the predictions at those points. In classification problems, ICT moves the decision boundary to low-density regions of the data distribution. Our experiments show that ICT achieves state-of-the-art performance when applied to standard neural network architectures on the CIFAR-10 and SVHN benchmark datasets. Our theoretical analysis shows that ICT corresponds to a certain type of data-adaptive regularization with unlabeled points which reduces overfitting to labeled points under high confidence values. (C) 2021 The Authors. Published by Elsevier Ltd.

关键词： Semi-supervised learning Deep Neural Networks Mixup Consistency regularization

来源：评论

学校读者我要写书评

暂无评论

Deep learning for high-resolution dose prediction in high dose rate brachytherapy for breast cancer treatment

引用

PHYSICS IN MEDICINE AND BIOLOGY 2024年第10期69卷 105011-105011页

作者： Quetin, Sebastien Bahoric, Boris Maleki, Farhad Enger, Shirin A. McGill Univ Dept Oncol Med Phys Unit Montreal PQ Canada Montreal Inst Learning Algorithms Mila Montreal PQ Canada McGill Univ Jewish Gen Hosp Dept Radiat Oncol Montreal PQ Canada Univ Calgary Dept Comp Sci Calgary AB Canada McGill Univ Dept Diagnost Radiol Montreal PQ Canada Univ Florida Dept Radiol Gainesville FL USA Jewish Gen Hosp Lady Davis Inst Med Res Montreal PQ Canada

Objective. Monte Carlo (MC) simulations are the benchmark for accurate radiotherapy dose calculations, notably in patient-specific high dose rate brachytherapy (HDR BT), in cases where considering tissue heterogeneities is critical. However, the lengthy computational time limits the practical application of MC simulations. Prior research used deep learning (DL) for dose prediction as an alternative to MC simulations. While accurate dose predictions akin to MC were attained, graphics processing unit limitations constrained these predictions to large voxels of 3 mm x 3 mm x 3 mm. This study aimed to enable dose predictions as accurate as MC simulations in 1 mm x 1 mm x 1 mm voxels within a clinically acceptable timeframe. Approach. Computed tomography scans of 98 breast cancer patients treated with Iridium-192-based HDR BT were used: 70 for training, 14 for validation, and 14 for testing. A new cropping strategy based on the distance to the seed was devised to reduce the volume size, enabling efficient training of 3D DL models using 1 mm x 1 mm x 1 mm dose grids. Additionally, novel DL architecture with layer-level fusion were proposed to predict MC simulated dose to medium-in-medium (D m,m ). These architectures fuse information from TG-43 dose to water-in-water (D w,w ) with patient tissue composition at the layer-level. Different inputs describing patient body composition were investigated. Main results. The proposed approach demonstrated state-of-the-art performance, on par with the MC D m,m maps, but 300 times faster. The mean absolute percent error for dosimetric indices between the MC and DL-predicted complete treatment plans was 0.17% +/- 0.15% for the planning target volume V 100, 0.30% +/- 0.32% for the skin D 2cc , 0.82% +/- 0.79% for the lung D 2cc , 0.34% +/- 0.29% for the chest wall D 2cc and 1.08% +/- 0.98% for the heart D 2cc . Significance. Unlike the time-consuming MC simulations, the proposed novel strategy efficiently converts TG-43 D w,w maps into

关键词： brachytherapy dose prediction deep learning Monte Carlo

来源：评论

学校读者我要写书评

暂无评论

Entropic Out-of-Distribution Detection: Seamless Detection of Unknown Examples

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND learning SYSTEMS 2022年第6期33卷 2350-2364页

作者： Macedo, David Ren, Tsang Ing Zanchettin, Cleber Oliveira, Adriano L. I. Ludermir, Teresa Univ Montreal UdeM Montreal Inst Learning Algorithms MILA Montreal PQ H3T 1J4 Canada Univ Fed Pernambuco Ctr Informat BR-50670901 Recife PE Brazil

In this article, we argue that the unsatisfactory out-of-distribution (OOD) detection performance of neural networks is mainly due to the SoftMax loss anisotropy and propensity to produce low entropy probability distributions in disagreement with the principle of maximum entropy. On the one hand, current OOD detection approaches usually do not directly fix the SoftMax loss drawbacks, but rather build techniques to circumvent it. Unfortunately, those methods usually produce undesired side effects (e.g., classification accuracy drop, additional hyperparameters, slower inferences, and collecting extra data). On the other hand, we propose replacing SoftMax loss with a novel loss function that does not suffer from the mentioned weaknesses. The proposed IsoMax loss is isotropic (exclusively distance-based) and provides high entropy posterior probability distributions. Replacing the SoftMax loss by IsoMax loss requires no model or training changes. Additionally, the models trained with IsoMax loss produce as fast and energy-efficient inferences as those trained using SoftMax loss. Moreover, no classification accuracy drop is observed. The proposed method does not rely on outlier/background data, hyperparameter tuning, temperature calibration, feature extraction, metric learning, adversarial training, ensemble procedures, or generative models. Our experiments showed that IsoMax loss works as a seamless SoftMax loss drop-in replacement that significantly improves neural networks' OOD detection performance. Hence, it may be used as a baseline OOD detection approach to be combined with current or future OOD detection techniques to achieve even higher results.

关键词： Entropy Training Neural networks Feature extraction Data preprocessing Measurement Prototypes Entropic score isotropy maximization loss maximum entropy principle out-of-distribution (OOD) detection

来源：评论

学校读者我要写书评

暂无评论

Interpolated Adversarial Training: Achieving robust neural networks without sacrificing too much accuracy

引用

NEURAL NETWORKS 2022年第0期154卷 218-233页

作者： Lamb, Alex Verma, Vikas Kawaguchi, Kenji Matyasko, Alexander Khosla, Savya Kannala, Juho Bengio, Yoshua Montreal Inst Learning Algorithms MILA Montreal PQ Canada Aalto Univ Espoo Finland Harvard Univ Cambridge MA 02138 USA Natl Univ Singapore Singapore Singapore Google New Delhi India

Adversarial robustness has become a central goal in deep learning, both in the theory and the practice. However, successful methods to improve the adversarial robustness (such as adversarial training) greatly hurt generalization performance on the unperturbed data. This could have a major impact on how the adversarial robustness affects real world systems (i.e. many may opt to forego robustness if it can improve accuracy on the unperturbed data). We propose Interpolated Adversarial Training, which employs recently proposed interpolation based training methods in the framework of adversarial training. On CIFAR-10, adversarial training increases the standard test error ( when there is no adversary) from 4.43% to 12.32%, whereas with our Interpolated adversarial training we retain the adversarial robustness while achieving a standard test error of only 6.45%. With our technique, the relative increase in the standard error for the robust model is reduced from 178.1% to just 45.5%. Moreover, we provide mathematical analysis of Interpolated Adversarial Training to confirm its efficiencies and demonstrate its advantages in terms of robustness and generalization. (C) 2022 The Authors. Published by Elsevier Ltd.

关键词： Adversarial robustness Mixup Manifold Mixup Standard test error

来源：评论

学校读者我要写书评

暂无评论

Distributed information-theoretic clustering

引用

INFORMATION AND INFERENCE-A JOURNAL OF THE IMA 2022年第1期11卷 137-166页

作者： Pichler, Georg Piantanida, Pablo Matz, Gerald TU Wien Inst Telecommun Vienna Austria Univ Paris Saclay CNRS Cent Supelec Lab Signaux & Syst Gif Sur Yvette France Univ Montreal Montreal Inst Learning Algorithms Mila Montreal PQ Canada

We study a novel multi-terminal source coding setup motivated by the biclustering problem. Two separate encoders observe two i.i.d. sequences X-n and Y-n, respectively. The goal is to find rate-limited encodings f (x(n)) and g(z(n)) that maximize the mutual information I(f (X-n);g(Y-n))/n. We discuss connections of this problem with hypothesis testing against independence, pattern recognition and the information bottleneck method. Improving previous cardinality bounds for the inner and outer bounds allows us to thoroughly study the special case of a binary symmetric source and to quantify the gap between the inner and the outer bound in this special case. Furthermore, we investigate a multiple description (MD) extension of the CEO problem with mutual information constraint. Surprisingly, thisMD-CEO problem permits a tight single-letter characterization of the achievable region.

关键词： source coding mutual information information bottleneck CEO problem

来源：评论

学校读者我要写书评

暂无评论

Conditional Generation of Medical Images via Disentangled Adversarial Inference 1st

Conditional Generation of Medical Images via Disentangled Ad...

引用

1st Workshop on Deep Generative Models for Medical Image Computing and Computer Assisted Intervention (DGM4MICCAI) / 1st MICCAI Workshop on Data Augmentation, Labelling, and Imperfections (DALI)

作者： Havaei, Mohammad Mao, Ximeng Wang, Yipping Lao, Qicheng Imagia Montreal PQ Canada Univ Montreal Montreal Inst Learning Algorithms MILA Montreal PQ Canada

ISBN: (纸本)9783030882105;9783030882099

We propose DRAI-a dual adversarial inference framework with augmented disentanglement constraints-to learn from the image itself, disentangled representations of style and content, and use this information to impose control over conditional generation process. We undergo two novel regularization steps to ensure content-style disentanglement. First, we minimize the shared information between content and style by introducing a novel application of the gradient reverse layer (GRL);second, we introduce a self-supervised regularization method to further separate information in the content and style variables. We conduct extensive qualitative and quantitative assessments on two publicly available medical imaging datasets (LIDC and HAM10000) and test for conditional image generation and style-content disentanglement. We also show that our proposed model (DRAI) achieves the best disentanglement score and has the best overall performance.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

Post-Editing Extractive Summaries by Definiteness Prediction

Post-Editing Extractive Summaries by Definiteness Prediction

引用

Meeting of the Association-for-Computational-Linguistics (ACL-EMNLP)

作者： Kabbara, Jad Cheung, Jackie Chi Kit McGill Univ Sch Comp Sci Montreal PQ Canada Montreal Inst Learning Algorithms Mila Montreal PQ Canada

ISBN: (纸本)9781955917100

Extractive summarization has been the mainstay of automatic summarization for decades. Despite all the progress, extractive summarizers still suffer from shortcomings including coreference issues arising from extracting sentences away from their original context in the source document. This affects the coherence and readability of extractive summaries. In this work, we propose a lightweight postediting step for extractive summaries that centers around a single linguistic decision: the definiteness of noun phrases. We conduct human evaluation studies that show that human expert judges substantially prefer the output of our proposed system over the original summaries. Moreover, based on an automatic evaluation study, we provide evidence for our system's ability to generate linguistic decisions that lead to improved extractive summaries. We also draw insights about how the automatic system is exploiting some local cues related to the writing style of the main article texts or summary texts to make the decisions, rather than reasoning about the contexts pragmatically.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：