1,685 Hits in 3.2 sec

Revisiting Deep Learning Models for Tabular Data [article]

Yury Gorishniy, Ivan Rubachev, Valentin Khrulkov, Artem Babenko
2021 arXiv   pre-print
The existing literature on deep learning for tabular data proposes a wide range of novel architectures and reports competitive results on various datasets.  ...  In this work, we perform an overview of the main families of DL architectures for tabular data and raise the bar of baselines in tabular DL by identifying two simple and powerful deep architectures.  ...  Conclusion In this work, we have investigated the status quo in the field of deep learning for tabular data and improved the state of baselines in tabular DL.  ... 
arXiv:2106.11959v2 fatcat:v4g4vvf3ojd25jkuyvjxwb26fy

Efficient Model-Based Deep Reinforcement Learning with Variational State Tabulation [article]

Dane Corneil, Wulfram Gerstner, Johanni Brea
2018 arXiv   pre-print
In this article we introduce Variational State Tabulation (VaST), which maps an environment with a high-dimensional state space (e.g. the space of visual inputs) to an abstract tabular model.  ...  To improve sample efficiency, an agent may build a model of the environment and use planning methods to update its policy.  ...  Acknowledgements We thank Vasiliki Liakoni and Marco Lehmann for their invaluable suggestions, feedback and corrections on the manuscript.  ... 
arXiv:1802.04325v2 fatcat:74up624o5be55hm462gzfcnklq

Solving reward-collecting problems with UAVs: a comparison of online optimization and Q-learning [article]

Yixuan Liu and Chrysafis Vogiatzis and Ruriko Yoshida and Erich Morman
2021 arXiv   pre-print
We present a comparison of three methods to solve this problem: namely we implement a Deep Q-Learning model, an ε-greedy tabular Q-Learning model, and an online optimization framework.  ...  As the prevalence of UAVs increases, there has also been improvements in counter-UAV technology that makes it difficult for them to successfully obtain valuable intelligence within an area of interest.  ...  Erich Morman: Modeled and implemented the ε-greedy tabular Q-Learning. Additionally conducted computational experiments using -Q learning.  ... 
arXiv:2112.00141v1 fatcat:motshvd4qrfvphe2hnoppyib4q

Well-tuned Simple Nets Excel on Tabular Datasets [article]

Arlind Kadra, Marius Lindauer, Frank Hutter, Josif Grabocka
2021 arXiv   pre-print
Tabular datasets are the last "unconquered castle" for deep learning, with traditional ML methods like Gradient-Boosted Decision Trees still performing strongly even against recent specialized neural architectures  ...  We empirically assess the impact of these regularization cocktails for MLPs in a large-scale empirical study comprising 40 tabular datasets and demonstrate that (i) well-regularized plain MLPs significantly  ...  deep learning on tabular data.  ... 
arXiv:2106.11189v2 fatcat:4ft4gdyvqfdohmmt4krfum3w64

Self-Consistent Models and Values [article]

Gregory Farquhar, Kate Baumli, Zita Marinho, Angelos Filos, Matteo Hessel, Hado van Hasselt, David Silver
2021 arXiv   pre-print
Learned models of the environment provide reinforcement learning (RL) agents with flexible ways of making predictions about the environment.  ...  In this work, we investigate a way of augmenting model-based RL, by additionally encouraging a learned model and value function to be jointly self-consistent.  ...  The authors received no specific funding for this work.  ... 
arXiv:2110.12840v1 fatcat:5ott7uqvavhodldt6nimv2ussu

YAHPO Gym – An Efficient Multi-Objective Multi-Fidelity Benchmark for Hyperparameter Optimization [article]

Florian Pfisterer, Lennart Schneider, Julia Moosbauer, Martin Binder, Bernd Bischl
2022 arXiv   pre-print
In this work, we propose a new set of challenging and relevant benchmark problems motivated by desirable properties and requirements for such benchmarks.  ...  Furthermore, we empirically compare surrogate-based benchmarks to the more widely-used tabular benchmarks, and demonstrate that the latter may produce unfaithful results regarding the performance ranking  ...  Acknowledgments The authors of this work take full responsibilities for its content. This work was supported by the German Federal Ministry of Education and Research (BMBF) under Grant No. 01IS18036A.  ... 
arXiv:2109.03670v3 fatcat:cggntj6uerhvrby6wjjwapyrba

Big Data in Preclinical ECG Alterations Research

Clemente Cipresso
2019 Biomedical Journal of Scientific & Technical Research  
A complete decryption of the electrocardiographic signal in numerical terms and the joint use of big data associated with modern Machine Learning systems will be able in the future to identify new parameters  ...  Same authors defined however the urgent need to revisit ECG reading mainly focusing research on data analysis and modelling [4] .  ...  In this context, the Machine e Deep Learning methodologies offer a great opportunity to achieve radical innovation in medical field related to cardiac diseases.  ... 
doi:10.26717/bjstr.2019.13.002384 fatcat:orjqc726dffz5mdel7kpik4w54

AutoGluon: A revolutionary framework for landslide hazard analysis

Wenwen Qi, Chong Xu, Xiwei Xu
2021 Natural Hazards Research  
It takes 47.33 seconds for data preprocessing and model 23 training for 11 machine learning models and the best result measured by Roc-AUC score is 0.94.  ...  Integrating machine learning models into landslide hazard 13 analysis is a common but challenging task for researchers in general.  ...  It 241 totally takes 47.33 seconds for data preprocessing and model training for 11 machine learning 242 models.  ... 
doi:10.1016/j.nhres.2021.07.002 fatcat:2ukopbcbgbdx3fg663zge42squ

Twin Systems for DeepCBR: A Menagerie of Deep Learning and Case-Based Reasoning Pairings for Explanation and Data Augmentation [article]

Mark T Keane and Eoin M Kenny and Mohammed Temraz and Derek Greene and Barry Smyth
2021 arXiv   pre-print
Recently, it has been proposed that fruitful synergies may exist between Deep Learning (DL) and Case Based Reasoning (CBR); that there are insights to be gained by applying CBR ideas to problems in DL  ...  We describe a series of twin-systems pairings of opaque DL models with transparent CBR models that allow the latter to explain the former using factual, counterfactual and semi-factual explanation strategies  ...  Data Augmentation Using Counterfactuals Thus far, we have seen that several productive links can be made between Deep Learning and CBR in the XAI field.  ... 
arXiv:2104.14461v2 fatcat:gj5zmgsob5hyjd5li3lbszoko4

Deep Weighted Averaging Classifiers [article]

Dallas Card and Michael Zhang and Noah A. Smith
2018 arXiv   pre-print
Recent advances in deep learning have achieved impressive gains in classification accuracy on a variety of types of data, including images and text.  ...  In this paper we propose a simple way to modify any conventional deep architecture to automatically provide more transparent explanations for classification decisions, as well as an intuitive notion of  ...  INTRODUCTION For any domain involving complex, structured, or high-dimensional data, deep learning has rapidly become the dominant approach for training classifiers.  ... 
arXiv:1811.02579v1 fatcat:tvsfvkkj2fcqxjsabq2txrsxwy

Organizing Experience: a Deeper Look at Replay Mechanisms for Sample-Based Planning in Continuous State Domains

Yangchen Pan, Muhammad Zaheer, Adam White, Andrew Patterson, Martha White
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
This elegant planning strategy has been mostly explored in the tabular setting. The aim of this paper is to revisit sample-based planning, in stochastic and continuous domains with learned models.  ...  Model-based strategies for control are critical to obtain sample efficient learning.  ...  We highlight criteria for learned models used within Dyna, and propose Reweighted Experience Models (REMs) that are data-efficient, efficient to sample and can be learned incrementally.  ... 
doi:10.24963/ijcai.2018/666 dblp:conf/ijcai/PanZWPW18 fatcat:evdqx3572fbfdnbckh4s555wjy

Scalable Coordinated Exploration in Concurrent Reinforcement Learning [article]

Maria Dimakopoulou, Ian Osband, Benjamin Van Roy
2018 arXiv   pre-print
We demonstrate that, for simple tabular contexts, the approach is competitive with previously proposed tabular model learning methods (Dimakopoulou and Van Roy, 2018).  ...  We consider a team of reinforcement learning agents that concurrently operate in a common environment, and we develop an approach to efficient coordinated exploration that is suitable for problems of practical  ...  are designed for tabular representations.  ... 
arXiv:1805.08948v2 fatcat:d7ew5vk47rcj3mhcncmf7pxpg4

Towards Tractable Optimism in Model-Based Reinforcement Learning [article]

Aldo Pacchiano and Philip J. Ball and Jack Parker-Holder and Krzysztof Choromanski and Stephen Roberts
2021 arXiv   pre-print
In the tabular setting, many state-of-the-art methods produce the required optimism through approaches which are intractable when scaling to deep RL.  ...  The principle of optimism in the face of uncertainty is prevalent throughout sequential decision making problems such as multi-armed bandits and reinforcement learning (RL).  ...  Playing atari with deep reinforcement model-based interval estimation for Markov decision pro- learning. In NIPS Deep Learning Workshop. 2013. cesses.  ... 
arXiv:2006.11911v2 fatcat:afhetowlfbdgra6cbiqx5qe6da

Propositionalization and Embeddings: Two Sides of the Same Coin [article]

Nada Lavrač and BlažŠkrlj and Marko Robnik-Šikonja
2020 arXiv   pre-print
While both approaches aim at transforming data into tabular data format, they use different terminology and task definitions, are perceived to address different goals, and are used in different contexts  ...  This paper contributes a unifying framework that allows for improved understanding of these two data transformation techniques by presenting their unified definitions, and by explaining the similarities  ...  Acknowledgements We acknowledge the financial support of the Slovenian Research Agency through core research programmes P2-0103 and P6-0411 and project Semantic Data Mining for Linked Open Data (financed  ... 
arXiv:2006.04410v1 fatcat:idpgnam52jdnbbpv32qhm7o3im

Propositionalization and embeddings: two sides of the same coin

Nada Lavrač, Blaž Škrlj, Marko Robnik-Šikonja
2020 Machine Learning  
While both approaches aim at transforming data into tabular data format, they use different terminology and task definitions, are perceived to address different goals, and are used in different contexts  ...  This paper contributes a unifying framework that allows for improved understanding of these two data transformation techniques by presenting their unified definitions, and by explaining the similarities  ...  The embedding technologies are mostly used in the context of deep learning for various data formats, including tabular data, texts, images, and graphs (including knowledge graphs).  ... 
doi:10.1007/s10994-020-05890-8 pmid:32704202 pmcid:PMC7366599 fatcat:byyvqrplkrdvbcqvfctswm3ncu
« Previous Showing results 1 — 15 out of 1,685 results