Filters








13,565 Hits in 3.7 sec

What Averages Do Not Tell – Predicting Real Life Processes with Sequential Deep Learning [article]

István Ketykó, Felix Mannhardt, Marwan Hassani, Boudewijn van Dongen
2021 arXiv   pre-print
Looking at suffix prediction as the most challenging of these tasks, the performance of Deep Learning models was evaluated only on average measures and for a small number of real-life event logs.  ...  Deep Learning is proven to be an effective tool for modeling sequential data as shown by the success in Natural Language, Computer Vision and Signal Processing.  ...  The performance of suffix prediction was evaluated only on average performance measures for a few real-life event logs [18] .  ... 
arXiv:2110.10225v2 fatcat:lvtsmlsdyff3nlvrgbu5ogrfce

Wikipedia for Smart Machines and Double Deep Machine Learning [article]

Moshe BenBassat
2018 arXiv   pre-print
Wikipedia for smart machines along with the new Double Deep Learning approach offer a paradigm for integrating datacentric deep learning algorithms with algorithms that leverage deep knowledge, e.g. evidential  ...  Following a review and illustrations of such limitations for several real life AI applications, we point at ways to overcome them.  ...  How are we going to represent such processes for AI? Can we achieve that with Deep learning algorithms?  ... 
arXiv:1711.06517v2 fatcat:qfroyn2pvncgjddtsa67srerfm

Churn Prediction with Sequential Data and Deep Neural Networks. A Comparative Analysis [article]

C. Gary Mena, Arno De Caigny, Kristof Coussement, Koen W. De Bock, Stefan Lessmann
2019 arXiv   pre-print
of such sequential data.  ...  Off-the-shelf machine learning algorithms for prediction such as regularized logistic regression cannot exploit the information of time-varying features without previously using an aggregation procedure  ...  Deep Neural NetworksArchitecture Layers Hidden Units Filter Size Optimizer Learning rate EpochsNotes: Algorithms are implemented with Keras using Tensorflow as backend. da = does not apply Batch size  ... 
arXiv:1909.11114v1 fatcat:qjuihguj3jfy3bgjtqkm6ajgbe

The Particle Track Reconstruction based on deep Neural networks

Dmitriy Baranov, Sergey Mitsyn, Pavel Goncharov, Gennady Ososkov, A. Forti, L. Betev, M. Litmaath, O. Smirnova, P. Hristov
2019 EPJ Web of Conferences  
We show that both proposed deep networks do not require any special preprocessing stage, are more accurate, faster and can be easier parallelized.  ...  One of the most important problems of data processing in high energy and nuclear physics is the event reconstruction.  ...  Also, we start predicting track hits based on the first point of the track and we suppose that every track starts from the first station, although in real life it is not always true.  ... 
doi:10.1051/epjconf/201921406018 fatcat:lgaoksgsyrem3faz2ifho6qi64

CCasGNN: Collaborative Cascade Prediction Based on Graph Neural Networks [article]

Yansong Wang, Xiaomeng Wang, Tao Jia
2021 arXiv   pre-print
The experiments conducted on two real-world datasets confirm that our method significantly improves the prediction accuracy compared to state-of-the-art approaches.  ...  Most previous methods concentrate on mining either structural or sequential features from the network and the propagation path.  ...  INTRODUCTION Social networks such as Twitter, Weibo, and YouTube have greatly facilitated our life, in which people post what they see and hear with friends [1] .  ... 
arXiv:2112.03644v1 fatcat:tbtn42xeizb6ricvcf72odk434

Attend and Predict: Understanding Gene Regulation by Selective Attention on Chromatin [article]

Ritambhara Singh, Jack Lanchantin, Arshdeep Sekhon, Yanjun Qi
2017 arXiv   pre-print
AttentiveChrome trains two levels of attention jointly with the target prediction, enabling it to attend differentially to relevant marks and to locate important positions per mark.  ...  Two fundamental challenges exist for such learning tasks: (1) genome-wide chromatin signals are spatially structured, high-dimensional and highly modular; and (2) the core aim is to understand what are  ...  We do this by learning a second level of attention among HMs.  ... 
arXiv:1708.00339v3 fatcat:jazef3yki5e3zhnvumdjabc2oe

Attend and Predict: Understanding Gene Regulation by Selective Attention on Chromatin

Ritambhara Singh, Jack Lanchantin, Arshdeep Sekhon, Yanjun Qi
2017 Advances in Neural Information Processing Systems  
AttentiveChrome trains two levels of attention jointly with the target prediction, enabling it to attend differentially to relevant marks and to locate important positions per mark.  ...  Two fundamental challenges exist for such learning tasks: (1) genome-wide chromatin signals are spatially structured, high-dimensional and highly modular; and (2) the core aim is to understand what the  ...  We do this by learning a second level of attention among HMs.  ... 
pmid:30147283 pmcid:PMC6105294 fatcat:nnsg3hfek5g4rmb5d4vwiszdfq

Attend and Predict: Understanding Gene Regulation by Selective Attention on Chromatin [article]

Ritambhara Singh, Jack Lanchantin, Arshdeep Sekhon, Yanjun Qi
2018 bioRxiv   pre-print
AttentiveChrome trains two levels of attention jointly with the target prediction, enabling it to attend differentially to relevant marks and to locate important positions per mark.  ...  Two fundamental challenges exist for such learning tasks: (1) genome-wide chromatin signals are spatially structured, high-dimensional and highly modular; and (2) the core aim is to understand what are  ...  Visually, we can tell that the average H prom profile is similar to H active .  ... 
doi:10.1101/329334 fatcat:dcd4oiwggbeerlextwaeugvlwu

SmartFall: A Smartwatch-Based Fall Detection System Using Deep Learning

Taylor Mauldin, Marc Canby, Vangelis Metsis, Anne Ngu, Coralys Rivera
2018 Sensors  
Furthermore, the Deep Learning model exhibits a better ability to generalize to new users when predicting falls, an important quality of any model that is to be successful in the real world.  ...  The smartwatch is paired with a smartphone that runs the SmartFall application, which performs the computation necessary for the prediction of falls in real time without incurring latency in communicating  ...  Manvick Paliwal and Po-Teng Tseng for helping with the fall data collection process. Conflicts of Interest: The authors declare no conflict of interest. Sensors 2018, 18, 3363  ... 
doi:10.3390/s18103363 pmid:30304768 pmcid:PMC6210545 fatcat:jm7jajbdabgajgvjwpjkefgakm

Deep reinforcement learning from human preferences [article]

Paul Christiano, Jan Leike, Tom B. Brown, Miljan Martic, Shane Legg, Dario Amodei
2017 arXiv   pre-print
For sophisticated reinforcement learning (RL) systems to interact usefully with real-world environments, we need to communicate complex goals to these systems.  ...  These behaviors and environments are considerably more complex than any that have been previously learned from human feedback.  ...  We replaced these termination conditions by a penalty which encourages the parameters to remain in the range (and which the agent must learn). • In Atari games, we do not send life loss or episode end  ... 
arXiv:1706.03741v3 fatcat:b2phuyaq7fay7chweuqdkbo4ae

Student Performance Prediction with Short-Term Sequential Campus Behaviors

Xinhua Wang, Xuemeng Yu, Lei Guo, Fangai Liu, Liancheng Xu
2020 Information  
process, which is an important step towards personalized education.  ...  Then,to conduct student performance prediction, we further involve these learned features to the classicSupport Vector Machine (SVM) algorithm and finally achieve our SPC model.  ...  Deep Learning-Based Sequence Modeling Deep learning-based sequence modeling aiming at capturing deep recurrent features from sequential items (such as words or products) has been widely studied in many  ... 
doi:10.3390/info11040201 fatcat:dljh2rsq3bfcrp3lwzawszdjdu

Augur

Ethan Fast, William McGrath, Pranav Rajpurkar, Michael S. Bernstein
2016 Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems - CHI '16  
meeting with a friend, or taking a selfie.  ...  From smart homes that prepare coffee when we wake, to phones that know not to interrupt us during important conversations, our collective visions of HCI imagine a future in which computers understand a  ...  However, if people type with computers far more often than anything else, then knowing there is a computer in your room tells us significant information, statistically, about what you might be doing.  ... 
doi:10.1145/2858036.2858528 dblp:conf/chi/FastMRB16 fatcat:3ebxaumhmff7zh4c7sb6xec274

Recommending Cryptocurrency Trading Points with Deep Reinforcement Learning Approach

Otabek Sattarov, Azamjon Muminov, Cheol Won Lee, Hyun Kyu Kang, Ryumduck Oh, Junho Ahn, Hyung Jun Oh, Heung Seok Jeon
2020 Applied Sciences  
To address this challenge, we tried to apply one of the machine learning algorithms, which is called deep reinforcement learning (DRL) on the stock market.  ...  As a result, we developed an application that observes historical price movements and takes action on real-time prices.  ...  For an explanation of how they are useful in real life, we downloaded bitcoin hourly historical price data and tested them with some of these strategies.  ... 
doi:10.3390/app10041506 fatcat:r6pysynusbdzpkhmcaczx2ihay

Stock Prediction Using Convolutional Neural Network

Sheng Chen, Hongxiang He
2018 IOP Conference Series: Materials Science and Engineering  
by using different ways now, including machine learning, deep learning and so on.  ...  In this paper, we proposed a deep learning method based on Convolutional Neural Network to predict the stock price movement of Chinese stock market.  ...  Generally speaking, as the data of stock usually can be seen as sequential data, the frequency of using Recurrent Neural Network to process the data related with time series are higher compared with using  ... 
doi:10.1088/1757-899x/435/1/012026 fatcat:a73kaxpb2nfjxixlrgqfn6lufu

Artificial intelligence driven resiliency with machine learning and deep learning components

Bahman Zohuri, Farhang Mossavar Rahmani
2020 Japan Journal of Research  
Artificial intelligence driven resiliency with machine learning and deep learning components. Japan J Res. 2019;1(1):1-5.  ...  RFID tags, sensors, and smart metering are driving the need to deal with torrents of data in near-real-time.  ...  However, it is not the amount of data that is important. It is what organizations do with the data that matters.  ... 
doi:10.33425/2690-8077.1002 fatcat:w2qrb7vmxrdjfp7igrxtre47gq
« Previous Showing results 1 — 15 out of 13,565 results