Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Iterative Feature Space Optimization through Incremental Adaptive Evaluation

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read EASE, a plug-in evaluator for iterative feature-space search, claims better selected features and lower cumulative training time than five baseline evaluators on fourteen datasets.

desk verdict A coherent plug-in evaluator with a novel architecture, but the headline performance gain isn't statistically supported and Eq. (3) needs fixing. read the letter →

arxiv 2501.14889 v2 pith:OQWV7AN4 submitted 2025-01-24 cs.LG

Yanping Wu , Yanyong Huang , Zhengzhang Chen , Zijun Yao , Yanjie Fu , Kunpeng Liu , Xiao Luo , Dongjie Wang This is my paper · ORCID
classification cs.LG
keywords featurespaceevaluationiterativeselectionincrementallearningelasticweightconsolidationmulti-headattentionfeature-samplesubspaceconstructionautomatedoptimization
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper proposes EASE, a plug-in evaluator for iterative feature-space optimization—the loop in which an algorithm repeatedly refines which features to use until a downstream predictor performs well. Its claim is that standard evaluators used inside such loops treat all samples alike, can overfit a chosen downstream model, and are retrained from scratch at every iteration. EASE addresses these problems by splitting the current feature space into feature-sample subspaces that emphasize the most task-relevant features and the hardest-to-evaluate samples, scoring the subspaces with a weight-sharing multi-head attention network, and updating that network incrementally with an elastic-weight-consolidation penalty instead of retraining. On fourteen datasets and inside four iterative feature-selection and feature-generation frameworks, the paper reports that EASE both improves the downstream performance of the resulting feature space and lowers cumulative optimization time relative to five baseline evaluators. If the claim holds, feature-space search—a common bottleneck in applied machine learning—could become faster and more model-agnostic.

What carries the argument

The load-bearing machinery is the Feature-Sample Subspace Generator together with the Contextual Attention Evaluator: one identifies which features matter and which samples are hardest, and the other encodes each subspace through multi-head attention with shared weights before producing a prediction. The Fisher information computed from the previous iteration's subspaces is what converts overlap between consecutive feature spaces into a training-time saving, via an elastic weight consolidation penalty that restrains changes to parameters judged important; this is what lets the evaluator be updated incrementally instead of retrained from scratch.

What would settle it

Run EASE and its retrain-from-scratch ablation on a dataset where consecutive feature-space candidates are made nearly disjoint, for example by using random restarts or reverse-order feature rankings, and compare cumulative training time and final downstream performance; if the incremental version no longer beats retraining from scratch, the overlap assumption is the load-bearing part. A second check is to compute the Fisher information on the first iteration's subspaces, then apply the same penalty to randomly selected parameters, which would test whether the penalty is specifically preserving task-relevant knowledge.

Watch

Extended reading notes

Core claim

EASE treats the evaluator itself as the thing to be optimized, not just the feature subset. Each iteration, a feature index optimizer ranks features by how much removing them changes the evaluator's loss, and a sample index optimizer samples instances proportionally to their previous prediction error; repeated draws over the selected indices form fixed-length feature subspaces. A contextual attention evaluator with weights shared across subspaces maps each subspace to a prediction, and an elastic weight consolidation term—a Fisher-information-weighted quadratic penalty on parameter movement from the previous iteration—lets the evaluator retain old knowledge while absorbing new information. The paper claims this combination makes the evaluator fairer, more generalizable, and cheaper, and reports that it outperforms linear/logistic regression, decision trees, GBDT, random forests, and XGBoost across the RFE, FLSR, SDAE, and GRFG frameworks and across classification and regression metrics.

Load-bearing premise

All of the reported speed and quality gains rest on the assumption that consecutive feature spaces in the optimization loop share enough information for the Fisher-information penalty to identify parameters worth preserving; if the feature space jumps too much between iterations, the penalty could freeze outdated knowledge and the efficiency gain would vanish.

Editorial extensions

If this is right

  • Swapping EASE into an existing iterative feature-selection loop should improve the downstream performance of the selected feature space, with reported classification gains around 3% and consistent regression gains on the test datasets.
  • Cumulative optimization time should drop substantially relative to tree-based and linear evaluators, with reported savings sometimes exceeding 100 seconds per dataset, because the evaluator no longer needs full retraining at each iteration.
  • The feature space selected by EASE transfers across downstream predictors, beating the original feature space when evaluated with LR, DT, GBDT, and RF.
  • EASE operates as a drop-in evaluator across RFE, FLSR, SDAE, and GRFG, so the benefit is not tied to a single search algorithm.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • An untested extension is to apply the same incremental-update design to other iterative loops that repeatedly re-score a moving object, such as neural architecture search or prompt optimization, where consecutive candidates share structure.
  • A testable prediction is that the efficiency gain degrades as consecutive feature spaces become less overlapping; varying the step size of feature removals between iterations would reveal how much of the gain comes from the overlap assumption.
  • A cleaner isolating experiment would compare EASE against a same-architecture attention evaluator retrained from scratch with the same subspace construction; the EASE-IT ablation moves in this direction but does not fully separate architecture gains from incremental-update gains.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes EASE, a plug-in feature-space evaluator for iterative feature-space optimization. EASE decouples a feature space into feature-sample subspaces by selecting important features and hard samples, scores the subspaces with a weight-sharing multi-head attention network, and updates the network incrementally using elastic weight consolidation (EWC). The authors integrate EASE into RFE, FLSR, SDAE, and GRFG and compare its downstream performance and cumulative training time against LR, DT, GBDT, RF, and XGB as alternative evaluators. The central claims are that EASE produces better feature spaces and reduces cumulative training time across iterative optimization frameworks.

Significance. If the empirical claims held, EASE would be a practically useful plug-in for reducing the cost of iterative feature selection/generation while improving downstream performance. The component design is sensible: the feature-sample subspace generator targets challenging samples and relevant features, and the EWC-based incremental update is a plausible mechanism for exploiting overlap between consecutive feature spaces. The paper also ships code (via an anonymous repository), evaluates with an external RF model in the testing phase, and includes ablations (EASE-PT, EASE-IT, EASE-FC) that isolate the contribution of each component. These are real strengths. However, the headline superiority claim is not statistically established, the feature-importance definition in Eq. (3) is a derivation gap, and the efficiency claim is only partially demonstrated. The contribution is plausible but needs stronger evidence before the central claims can be accepted.

major comments (4)
  1. [5.2.1, Tables 2 and 5; Appendix C.1] The central performance claim is not statistically supported. All experiments are run only 10 times (Appendix C.1) and no significance tests or confidence intervals on differences are reported. Many decisive entries have overlapping standard deviations, e.g., Table 2 openml_616 R2: EASE 0.840±0.035 vs XGB 0.837±0.016, and Table 6 svmguide3 Accuracy: EASE 0.856±0.014 vs RF 0.861±0.012, where a baseline is numerically higher. Without paired tests (e.g., Wilcoxon signed-rank) or confidence intervals on per-dataset differences, the statement that 'EASE outperforms other baselines in most cases' (Section 5.2.1) is not established. This is load-bearing for the abstract and the paper's main contribution, so the authors should add appropriate statistical evidence or soften the claim.
  2. [4.1, Eq. (3)] The feature importance score is defined as Score(f_i) = M(F(t); Theta) - M(F(t) \ {f_i}; Theta), where M is the evaluator that produces predictions according to Eqs. (1)-(2). Subtracting raw evaluator outputs is not the same as measuring the impact on model performance; a feature can change predictions without changing loss or accuracy, and vice versa. The text in Section 4.1 says the score assesses 'the impact of removing that feature on the performance of the model,' but Eq. (3) does not implement that. Please either redefine M to output a scalar performance estimate (e.g., expected loss), or present Eq. (3) explicitly as a sensitivity heuristic and validate that it tracks actual performance differences.
  3. [5.2.2, Figures 3 and 6; Appendix C.3] The efficiency claim is only demonstrated for the RFE framework and only against GBDT, RF, and XGB. No cumulative-time comparison against LR or DT is reported, and no cumulative-time comparison is given inside FLSR, SDAE, or GRFG, even though the paper claims that EASE 'can efficiently assess feature space quality' in general and highlights efficiency as a key motivation (Section 1, Section 5.2.2). The reported speed-ups in Figures 3 and 6 are therefore specific to RFE with three baselines. Please add time comparisons in the other frameworks, or explicitly scope the efficiency claim to RFE and the compared baselines.
  4. [4.3, Eqs. (10)-(11)] The EWC incremental update assumes that the Fisher information computed from the previous iteration's subspaces, Eq. (10), correctly identifies the parameters that remain important for evaluating the current feature space. The manuscript asserts the informational overlap shown in Figure 1b, but it never tests the assumption under low-overlap conditions, e.g., when consecutive feature spaces share few features or when the optimizer makes a large change. If the overlap assumption fails, the EWC penalty may freeze outdated parameters and the reported efficiency gain would disappear. A controlled experiment varying the degree of overlap, or at least a discussion of when the assumption is violated, is needed to support the mechanism behind the claimed speed-up.
minor comments (5)
  1. [Table 1 and Abstract] Table 1 lists 'spam_base' twice, and the abstract claims 'fourteen real-world datasets' while the introduction and Section 5.1 say 'twelve' (six classification and six regression tasks). Please reconcile the counts and remove the duplicate row.
  2. [5.1, Baseline Algorithms] The paragraph says 'We apply EASE to two iterative feature selection frameworks' but then enumerates four methods: RFE, FLSR, SDAE, and GRFG. Please correct the count or restructure the sentence.
  3. [5.2.4 and C.6] There are small language errors: 'We Additionally test' should be 'We additionally test' in Section 5.2.4, and 'EASE −F Cby' should be 'EASE −F C by' in Appendix C.6.
  4. [Figures 3 and 6] The y-axis label 'Cumulative Time' has no units; Table 7 reports seconds, so please state the unit in the figures for reproducibility.
  5. [5.2.1] The sentence 'For classification, EASE can improve by approximately 3% compared to other baselines' is vague; please specify whether this is an average accuracy gain, over which baselines, and for which metric.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: downstream claims are evaluated with an external RF model, and self-citations serve only as test beds or related work.

full rationale

EASE's two central claims are that it produces feature spaces with better downstream performance and that it does so more efficiently. Neither claim reduces to its own construction. The downstream evaluation in Section 5.1 is explicitly performed with an external model: 'In the testing phase, we use RF in all cases to report the performance of the refined feature space, as the model is stable and helps mitigate bias caused by the downstream model.' Thus the reported gains in Tables 2, 5, 6, and 8 are measured by a model outside EASE's parameters, not by EASE's own scoring function. The internal feedback loop in Eq. (3) and the sample-index error sampling is an algorithmic mechanism, not a prediction claim; it does not define the reported outcome. The incremental EWC update in Eq. (11) is a training procedure, and its efficiency gain is an empirical assertion, not a tautology. Self-citations to FLSR, GRFG, and prior feature-optimization work are used as test beds or related work, not as the authority for EASE's superiority. No fitted parameter is renamed as a prediction, and no uniqueness theorem is imported. The absence of significance tests in Tables 2, 5, 6, and 8 is a statistical-support weakness, not a circularity. The paper is therefore substantially self-contained against external benchmarks; only minor non-load-bearing self-citations justify a score slightly above zero.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

The claim rests on standard supervised learning assumptions plus the design choices encoded in the free parameters. The main added burden beyond normal hyperparameter tuning is the assumption that EWC transfer works across feature space iterations, and that the evaluator's own error is a good guide for hard-sample selection. No new physical entities or mathematical axioms are introduced.

free parameters (7)
  • k (number of top features selected per iteration) = not reported
    Feature Index Optimizer selects top-k features by importance score; k is not given in the paper.
  • q (number of feature subspaces) = not reported
    Feature Subspace Construction samples s indices q times; q is unspecified.
  • s (subspace sample size) = not reported
    Each subspace B_i has s samples; s is not reported.
  • lambda (EWC regularization coefficient) = not reported
    Eq. (11) uses lambda to balance new and old knowledge; the value is not given.
  • D (embedding/attention dimension) = 32 (used in sensitivity analysis)
    Sensitivity analysis fixes D=32 while varying heads, but the default for all experiments is not stated.
  • h (number of attention heads) = 16 (used in sensitivity analysis)
    Sensitivity analysis fixes h=16 while varying D, but the default for all experiments is not stated.
  • T (number of optimization iterations) = not reported
    Stopping criterion says maximum iterations, but the value is not given.
assumptions (5)
  • domain assumption The evaluator's prediction error on a sample is a valid proxy for that sample's difficulty or informativeness for evaluating feature space quality.
    Sample Index Optimizer samples rows with probability proportional to previous prediction error (Section 4.1).
  • domain assumption Consecutive feature spaces share enough overlapping information that the Fisher information from iteration t-1 identifies parameters important for iteration t.
    Incremental update relies on EWC; Section 4.3, Eq. (10)-(11), Figure 1b.
  • domain assumption The weight-sharing multi-head attention evaluator can capture feature-feature interactions that are relevant to downstream model performance.
    Section 4.2 states this as the motivation; no proof or analysis is given.
  • domain assumption Removing a feature and measuring the change in the evaluator's output (Eq. 3) provides a reliable ranking of feature importance.
    Feature Index Optimizer, Section 4.1, Eq. (3).
  • domain assumption The train/test evaluation protocol is a standard i.i.d. split; the paper reports test performance on a held-out set.
    Section 5.1 describes datasets and metrics but does not describe how data is split or whether cross-validation is used.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Iterative Feature Space Optimization through Incremental Adaptive Evaluation." pith.science (2026). https://pith.science/paper/OQWV7AN4

@misc{pith2026250114889,
  author       = {Pith},
  title        = {Pith review of: Iterative Feature Space Optimization through Incremental Adaptive Evaluation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OQWV7AN4}},
  note         = {Machine review of arXiv:2501.14889}
}
read the original abstract

Iterative feature space optimization involves systematically evaluating and adjusting the feature space to improve downstream task performance. However, existing works suffer from three key limitations:1) overlooking differences among data samples leads to evaluation bias; 2) tailoring feature spaces to specific machine learning models results in overfitting and poor generalization; 3) requiring the evaluator to be retrained from scratch during each optimization iteration significantly reduces the overall efficiency of the optimization process. To bridge these gaps, we propose a gEneralized Adaptive feature Space Evaluator (EASE) to efficiently produce optimal and generalized feature spaces. This framework consists of two key components: Feature-Sample Subspace Generator and Contextual Attention Evaluator. The first component aims to decouple the information distribution within the feature space to mitigate evaluation bias. To achieve this, we first identify features most relevant to prediction tasks and samples most challenging for evaluation based on feedback from the subsequent evaluator. This decoupling strategy makes the evaluator consistently target the most challenging aspects of the feature space. The second component intends to incrementally capture evolving patterns of the feature space for efficient evaluation. We propose a weighted-sharing multi-head attention mechanism to encode key characteristics of the feature space into an embedding vector for evaluation. Moreover, the evaluator is updated incrementally, retaining prior evaluation knowledge while incorporating new insights, as consecutive feature spaces during the optimization process share partial information. Extensive experiments on fourteen real-world datasets demonstrate the effectiveness of the proposed framework. Our code and data are publicly available.

Figures

Figures reproduced from arXiv: 2501.14889 by the authors.

Figure 1
Figure 1. (a) Illustration of the iterative feature space optimization, where the optimization module refines the feature [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Framework overview and parameter update for [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Time complexity comparison of different feature space evaluators across various datasets. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Comparison of prediction performance between original and EASE refined feature spaces. [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Comparison of refinement performance of feature space evaluators within FLSR. [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: Time complexity comparison of different feature space evaluators across various datasets. [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 7
Figure 7. Figure 7: Comparison of prediction performance between original and EASE refined feature spaces. [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]
Figure 8
Figure 8. Figure 8: Parameter sensitivity on the number of heads [PITH_FULL_IMAGE:figures/full_fig_p017_8.png]
Figure 9
Figure 9. Figure 9: Comparison of feature importance in EASE feature space and original feature space. [PITH_FULL_IMAGE:figures/full_fig_p018_9.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. GPT-FT: An Efficient Automated Feature Transformation Using GPT for Sequence Reconstruction and Performance Enhancement

    cs.LG 2025-08 conditional novelty 4.0 of 10

    GPT-FT replaces the LSTM encoder-decoder of MOAT with a small decoder-only GPT that both reconstructs transformation sequences and predicts their performance, enabling faster gradient-based feature search.

Reference graph

Works this paper leans on

48 extracted references · 40 canonical work pages · cited by 1 Pith paper

  1. [1]

    Feature dimensionality reduction: a review

    Weikuan Jia, Meili Sun, Jian Lian, and Sujuan Hou. Feature dimensionality reduction: a review. Complex & Intelligent Systems, 8 0 (3): 0 2663--2693, 2022

    work page 2022

  2. [2]

    A hybrid artificial immune optimization for high-dimensional feature selection

    Yongbin Zhu, Wenshan Li, and Tao Li. A hybrid artificial immune optimization for high-dimensional feature selection. Knowledge-Based Systems, 260: 0 110111, 2023

    work page 2023

  3. [3]

    Survey of feature selection and extraction techniques for stock market prediction

    Htet Htet Htun, Michael Biehl, and Nicolai Petkov. Survey of feature selection and extraction techniques for stock market prediction. Financial Innovation, 9 0 (1): 0 26, 2023

    work page 2023

  4. [4]

    A hybrid filter-wrapper feature selection using fuzzy knn based on bonferroni mean for medical datasets classification: A covid-19 case study

    Amukta Malyada Vommi and Tirumala Krishna Battula. A hybrid filter-wrapper feature selection using fuzzy knn based on bonferroni mean for medical datasets classification: A covid-19 case study. Expert Systems with Applications, 218: 0 119612, 2023

    work page 2023

  5. [5]

    A comprehensive review of dimensionality reduction techniques for feature selection and feature extraction

    Rizgar Zebari, Adnan Abdulazeez, Diyar Zeebaree, Dilovan Zebari, and Jwan Saeed. A comprehensive review of dimensionality reduction techniques for feature selection and feature extraction. Journal of Applied Science and Technology Trends, 1 0 (1): 0 56--70, 2020

    work page 2020

  6. [6]

    Using recursive feature elimination in random forest to account for correlated variables in high dimensional data

    Burcu F Darst, Kristen C Malecki, and Corinne D Engelman. Using recursive feature elimination in random forest to account for correlated variables in high dimensional data. BMC genetics, 19: 0 1--6, 2018

    work page 2018

  7. [7]

    A systematic literature review: Recursive feature elimination algorithms

    Arif Mudi Priyatno, Triyanna Widiyaningtyas, et al. A systematic literature review: Recursive feature elimination algorithms. JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer), 9 0 (2): 0 196--207, 2024

    work page 2024

  8. [8]

    Recursive feature elimination by sensitivity testing

    Nicholas Sean Escanilla, Lisa Hellerstein, Ross Kleiman, Zhaobin Kuang, James Shull, and David Page. Recursive feature elimination by sensitivity testing. In 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA), pages 40--47, 2018. doi:10.1109/ICMLA.2018.00014

    arXiv 2018

Show all 48 references
  1. [9]

    Traceable group-wise self-optimizing feature transformation learning: A dual optimization perspective

    Meng Xiao, Dongjie Wang, Min Wu, Kunpeng Liu, Hui Xiong, Yuanchun Zhou, and Yanjie Fu. Traceable group-wise self-optimizing feature transformation learning: A dual optimization perspective. ACM Transactions on Knowledge Discovery from Data, 18 0 (4): 0 1--22, 2024

    2024

  2. [10]

    Mel: efficient multi-task evolutionary learning for high-dimensional feature selection

    Xubin Wang, Haojiong Shangguan, Fengyi Huang, Shangrui Wu, and Weijia Jia. Mel: efficient multi-task evolutionary learning for high-dimensional feature selection. IEEE Transactions on Knowledge and Data Engineering, 2024

    2024

  3. [11]

    Automated feature selection: A reinforcement learning perspective

    Kunpeng Liu, Yanjie Fu, Le Wu, Xiaolin Li, Charu Aggarwal, and Hui Xiong. Automated feature selection: A reinforcement learning perspective. IEEE Transactions on Knowledge and Data Engineering, 35 0 (3): 0 2272--2284, 2021 a

    2021

  4. [12]

    Group-wise reinforcement feature generation for optimal and explainable representation space reconstruction

    Dongjie Wang, Yanjie Fu, Kunpeng Liu, Xiaolin Li, and Yan Solihin. Group-wise reinforcement feature generation for optimal and explainable representation space reconstruction. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 1826--...

    2022

  5. [13]

    Class-incremental learning via dual augmentation

    Fei Zhu, Zhen Cheng, Xu-Yao Zhang, and Cheng-lin Liu. Class-incremental learning via dual augmentation. Advances in Neural Information Processing Systems, 34: 0 14306--14318, 2021

    2021

  6. [14]

    Continual learning strategy in one-stage object detection framework based on experience replay for autonomous driving vehicle

    Jeng-Lun Shieh, Qazi Mazhar ul Haq, Muhamad Amirul Haq, Said Karam, Peter Chondro, De-Qin Gao, and Shanq-Jang Ruan. Continual learning strategy in one-stage object detection framework based on experience replay for autonomous driving vehicle. Sensors, 20 0 (23): 0 6777, 2020

    2020

  7. [15]

    Batch-incremental versus instance-incremental learning in dynamic and evolving data

    Jesse Read, Albert Bifet, Bernhard Pfahringer, and Geoff Holmes. Batch-incremental versus instance-incremental learning in dynamic and evolving data. In Advances in Intelligent Data Analysis XI: 11th International Symposium, IDA 2012, Helsinki, Finland, October 25-27, 2012. Pr...

    2012

  8. [16]

    Online class-incremental continual learning with adversarial shapley value

    Dongsub Shim, Zheda Mai, Jihwan Jeong, Scott Sanner, Hyunwoo Kim, and Jongseong Jang. Online class-incremental continual learning with adversarial shapley value. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 9630--9638, 2021

    2021

  9. [17]

    Overcoming catastrophic forgetting in neural networks

    James Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, et al. Overcoming catastrophic forgetting in neural networks. Proceedings of the national academy of scien...

    2017

  10. [18]

    Learning without forgetting

    Zhizhong Li and Derek Hoiem. Learning without forgetting. IEEE transactions on pattern analysis and machine intelligence, 40 0 (12): 0 2935--2947, 2017

    2017

  11. [19]

    Selective experience replay for lifelong learning

    David Isele and Akansel Cosgun. Selective experience replay for lifelong learning. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 32, 2018

    2018

  12. [20]

    Continual learning with deep generative replay

    Hanul Shin, Jung Kwon Lee, Jaehong Kim, and Jiwon Kim. Continual learning with deep generative replay. Advances in neural information processing systems, 30, 2017

    2017

  13. [21]

    Random path selection for continual learning

    Jathushan Rajasegaran, Munawar Hayat, Salman H Khan, Fahad Shahbaz Khan, and Ling Shao. Random path selection for continual learning. Advances in neural information processing systems, 32, 2019

    2019

  14. [22]

    Overcoming catastrophic forgetting with hard attention to the task

    Joan Serra, Didac Suris, Marius Miron, and Alexandros Karatzoglou. Overcoming catastrophic forgetting with hard attention to the task. In International conference on machine learning, pages 4548--4557. PMLR, 2018

    2018

  15. [23]

    Progressive neural network-based knowledge transfer in acoustic models

    Takafumi Moriya, Ryo Masumura, Taichi Asami, Yusuke Shinohara, Marc Delcroix, Yoshikazu Yamaguchi, and Yushi Aono. Progressive neural network-based knowledge transfer in acoustic models. In 2018 Asia-Pacific Signal and Information Processing Association Annual Summit and Confe...

    2018

  16. [24]

    Expert gate: Lifelong learning with a network of experts

    Rahaf Aljundi, Punarjay Chakravarty, and Tinne Tuytelaars. Expert gate: Lifelong learning with a network of experts. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3366--3375, 2017

    2017

  17. [25]

    L. Liu, Z. Kuang, Y. Chen, Jh. Xue, W. Yang, and W. Zhang. Incdet: In defense of elastic weight consolidation for incremental object detection. IEEE transactions on neural networks and learning systems, 32 0 (6): 0 2306--2319, 2021 b

    2021

  18. [26]

    Attention is all you need

    A Vaswani. Attention is all you need. Advances in Neural Information Processing Systems, 2017

    2017

  19. [27]

    Trajectory prediction for autonomous driving based on multi-head attention with joint agent-map representation

    Kaouther Messaoud, Nachiket Deo, Mohan M Trivedi, and Fawzi Nashashibi. Trajectory prediction for autonomous driving based on multi-head attention with joint agent-map representation. In 2021 IEEE Intelligent Vehicles Symposium (IV), pages 165--170. IEEE, 2021

    2021

  20. [28]

    Generating diverse translation by manipulating multi-head attention

    Zewei Sun, Shujian Huang, Hao-Ran Wei, Xin-yu Dai, and Jiajun Chen. Generating diverse translation by manipulating multi-head attention. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 8976--8983, 2020

    2020

  21. [29]

    Dynamic head: Unifying object detection heads with attentions

    Xiyang Dai, Yinpeng Chen, Bin Xiao, Dongdong Chen, Mengchen Liu, Lu Yuan, and Lei Zhang. Dynamic head: Unifying object detection heads with attentions. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 7373--7382, June 2021

    2021

  22. [30]

    A survey on swarm intelligence approaches to feature selection in data mining

    Bach Hoai Nguyen, Bing Xue, and Mengjie Zhang. A survey on swarm intelligence approaches to feature selection in data mining. Swarm and Evolutionary Computation, 54: 0 100663, 2020

    2020

  23. [31]

    A review of feature selection methods for machine learning-based disease risk prediction

    Nicholas Pudjihartono, Tayaza Fadason, Andreas W Kempa-Liehr, and Justin M O'Sullivan. A review of feature selection methods for machine learning-based disease risk prediction. Frontiers in Bioinformatics, 2: 0 927312, 2022

    2022

  24. [32]

    A bolasso based consistent feature selection enabled random forest classification algorithm: An application to credit risk assessment

    Nisha Arora and Pankaj Deep Kaur. A bolasso based consistent feature selection enabled random forest classification algorithm: An application to credit risk assessment. Applied Soft Computing, 86: 0 105936, 2020

    2020

  25. [33]

    A novel multi-objective forest optimization algorithm for wrapper feature selection

    Babak Nouri-Moghaddam, Mehdi Ghazanfari, and Mohammad Fathian. A novel multi-objective forest optimization algorithm for wrapper feature selection. Expert Systems with Applications, 175: 0 114737, 2021

    2021

  26. [34]

    A novel relation aware wrapper method for feature selection

    Zhaogeng Liu, Jielong Yang, Li Wang, and Yi Chang. A novel relation aware wrapper method for feature selection. Pattern Recognition, 140: 0 109566, 2023

    2023

  27. [35]

    Gene selection for cancer classification using support vector machines

    Isabelle Guyon, Jason Weston, Stephen Barnhill, and Vladimir Vapnik. Gene selection for cancer classification using support vector machines. Machine learning, 46: 0 389--422, 2002

    2002

  28. [36]

    A kronecker-factored approximate fisher matrix for convolution layers

    Roger Grosse and James Martens. A kronecker-factored approximate fisher matrix for convolution layers. In International Conference on Machine Learning, pages 573--582. PMLR, 2016

    2016

  29. [37]

    Openml dataset download

    Public. Openml dataset download. [EB/OL]. https://www.openml.org, 2024 a

    2024

  30. [38]

    Uci dataset download

    Public. Uci dataset download. [EB/OL]. https://archive.ics.uci.edu/, 2024 b

    2024

  31. [39]

    Kaggle dataset download

    Public. Kaggle dataset download. [EB/OL]. https://www.kaggle.com/c/amazon-employee-access-challenge/data, 2024 c

    2024

  32. [40]

    Simplifying reinforced feature selection via restructured choice strategy of single agent

    Xiaosa Zhao, Kunpeng Liu, Wei Fan, Lu Jiang, Xiaowei Zhao, Minghao Yin, and Yanjie Fu. Simplifying reinforced feature selection via restructured choice strategy of single agent. In 2020 IEEE International conference on data mining (ICDM), pages 871--880. IEEE, 2020

    2020

  33. [41]

    Selective deep autoencoder for unsupervised feature selection

    Wael Hassanieh and Abdallah Chehade. Selective deep autoencoder for unsupervised feature selection. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 12322--12330, 2024

    2024

  34. [42]

    Linear regression

    Xiaogang Su, Xin Yan, and Chih-Ling Tsai. Linear regression. Wiley Interdisciplinary Reviews: Computational Statistics, 4 0 (3): 0 275--294, 2012

    2012

  35. [43]

    Logistic regression was as good as machine learning for predicting major chronic diseases

    Simon Nusinovici, Yih Chung Tham, Marco Yu Chak Yan, Daniel Shu Wei Ting, Jialiang Li, Charumathi Sabanayagam, Tien Yin Wong, and Ching-Yu Cheng. Logistic regression was as good as machine learning for predicting major chronic diseases. Journal of clinical epidemiology, 122: 0...

    2020

  36. [44]

    Predicting restaurant financial distress using decision tree and adaboosted decision tree models

    Soo Y Kim and Arun Upneja. Predicting restaurant financial distress using decision tree and adaboosted decision tree models. Economic Modelling, 36: 0 354--362, 2014

    2014

  37. [45]

    Interpolation of gnss position time series using gbdt, xgboost, and rf machine learning algorithms and models error analysis

    Zhen Li, Tieding Lu, Kegen Yu, and Jie Wang. Interpolation of gnss position time series using gbdt, xgboost, and rf machine learning algorithms and models error analysis. Remote Sensing, 15 0 (18): 0 4374, 2023

    2023

  38. [46]

    Predicting the carbon dioxide emission caused by road transport using a random forest (rf) model combined by meta-heuristic algorithms

    Hamed Khajavi and Amir Rastgoo. Predicting the carbon dioxide emission caused by road transport using a random forest (rf) model combined by meta-heuristic algorithms. Sustainable Cities and Society, 93: 0 104503, 2023

    2023

  39. [47]

    Enhancing the prediction of student performance based on the machine learning xgboost algorithm

    Amal Asselman, Mohamed Khaldi, and Souhaib Aammou. Enhancing the prediction of student performance based on the machine learning xgboost algorithm. Interactive Learning Environments, 31 0 (6): 0 3360--3379, 2023

    2023

  40. [48]

    Interpretable deep learning framework for land use and land cover classification in remote sensing using shap

    Anastasios Temenos, Nikos Temenos, Maria Kaselimi, Anastasios Doulamis, and Nikolaos Doulamis. Interpretable deep learning framework for land use and land cover classification in remote sensing using shap. IEEE Geoscience and Remote Sensing Letters, 20: 0 1--5, 2023

    2023

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.