Filters








904,529 Hits in 4.7 sec

Model-Based Deep Learning [article]

Nir Shlezinger, Jay Whang, Yonina C. Eldar, Alexandros G. Dimakis
2021 arXiv   pre-print
In this article we survey the leading approaches for studying and designing model-based deep learning systems.  ...  Such model-based deep learning methods exploit both partial domain knowledge, via mathematical structures designed for specific problems, as well as learning from limited data.  ...  Some representative issues and their relationship with the recommended model-based deep learning approaches include: 1) Missing domain knowledge -model-based deep learning can implement the model-based  ... 
arXiv:2012.08405v2 fatcat:4ilqi3vv4rar5gsveqzo4loqpy

Multi-task learning with deep model based reinforcement learning [article]

Asier Mujika
2017 arXiv   pre-print
In this paper, we present a model based approach to deep reinforcement learning which we use to solve different tasks simultaneously.  ...  In recent years, model-free methods that use deep learning have achieved great success in many different reinforcement learning environments.  ...  DISCUSSION We have presented a novel model based approach to deep reinforcement learning that opens new lines of research in this area.  ... 
arXiv:1611.01457v4 fatcat:l7bdberxpze3rej4asioliwa6u

Learning to Paint With Model-based Deep Reinforcement Learning [article]

Zhewei Huang, Wen Heng, Shuchang Zhou
2019 arXiv   pre-print
By employing a neural renderer in model-based Deep Reinforcement Learning (DRL), our agents learn to determine the position and color of each stroke and make long-term plans to decompose texture-rich images  ...  The training is based on the Deep Reinforcement Learning framework, which encourages the agent to make long-term plans for sequential stroke-based painting.  ...  Learning In this section, we introduce how to train the agent using the model-based DDPG algorithm.  ... 
arXiv:1903.04411v3 fatcat:j5dvpxwanzc4vda3aue3j4p6t4

Model-Based Deep Learning: On the Intersection of Deep Learning and Optimization [article]

Nir Shlezinger, Yonina C. Eldar, Stephen P. Boyd
2022 arXiv   pre-print
Model-based optimization and data-centric deep learning are often considered to be distinct disciplines.  ...  deep learning.  ...  form of model-based deep learning.  ... 
arXiv:2205.02640v2 fatcat:yclu5hqsx5bv7k2fg4gnfhyrma

Learning to Fly via Deep Model-Based Reinforcement Learning [article]

Philip Becker-Ehmck, Maximilian Karl, Jan Peters, Patrick van der Smagt
2020 arXiv   pre-print
In this work, by leveraging a learnt probabilistic model of drone dynamics, we learn a thrust-attitude controller for a quadrotor through model-based reinforcement learning.  ...  Learning to control robots without requiring engineered models has been a long-term goal, promising diverse and novel applications.  ...  Reinforcement Learning Model-free deep RL has received a tremendous amount of attention ever since Q-learning was successfully applied to playing Atari games directly from raw input images with the use  ... 
arXiv:2003.08876v3 fatcat:gonirfi77rdvxbditv2pnop5lu

Deep Learning Based Vehicle Make-Model Classification [article]

Burak Satar, Ahmet Emir Dirik
2018 arXiv   pre-print
Then, we feed them into the CNN model. It is reached approximately 4% better classification accuracy result than using a conventional CNN model.  ...  A pipeline is proposed to combine an SSD (Single Shot Multibox Detector) model with a CNN (Convolutional Neural Network) model to train on the database.  ...  Fig. 3 : 3 The architecture of the SSD model for detection Fig. 5 : 5 Overall accuracy result of Experiment III, VGG based weights of SSDWe also tested the SSD based model on some videos.  ... 
arXiv:1809.00953v1 fatcat:olj2pzzrsjgmrijhfh5ssnre6e

MoDL: Model Based Deep Learning Architecture for Inverse Problems [article]

Hemant Kumar Aggarwal, Merry P. Mani, Mathews Jacob
2018 arXiv   pre-print
We introduce a model-based image reconstruction framework with a convolution neural network (CNN) based regularization prior.  ...  Since the forward model is explicitly accounted for, a smaller network with fewer parameters is sufficient to capture the image information compared to black-box deep learning approaches, thus reducing  ...  The proposed framework, termed as MOdel-based reconstruction using Deep Learned priors (MoDL), merges the power of model-based reconstruction schemes with deep learning.  ... 
arXiv:1712.02862v3 fatcat:es47nwox2baktgrqpujm5mrura

Poisoning Deep Learning Based Recommender Model in Federated Learning Scenarios [article]

Dazhong Rong, Qinming He, Jianhai Chen
2022 arXiv   pre-print
For proving current federated recommendation is still vulnerable, in this work we probe to design attack approaches targeting deep learning based recommender models in federated learning scenarios.  ...  Extensive experiments show that our well-designed attacks can effectively poison the target models, and the attack effectiveness sets the state-of-the-art.  ...  We aim to poison deep learning based recommender model in FL scenarios without the prior knowledge.  ... 
arXiv:2204.13594v2 fatcat:qki4v777t5hhxfhxvqow4sb62i

Explaining Deep Learning-Based Driver Models

Maria Paz Sesmero Lorente, Elena Magán Lopez, Laura Alvarez Florez, Agapito Ledezma Espino, José Antonio Iglesias Martínez, Araceli Sanchis de Miguel
2021 Applied Sciences  
Different systems based on Artificial Intelligence (AI) techniques are currently used in relevant areas such as healthcare, cybersecurity, natural language processing, and self-driving cars.  ...  The proposed model is based on the cumulative prospect theory (CPT) [49] , and the model parameters are learned using a hierarchical learning algorithm based on inverse reinforcement learning [50] and  ...  In [34] , a deep facial expression recognition algorithm for emotions based on CNNs and an ensemble deep learning algorithm to predict facial expressions are proposed.  ... 
doi:10.3390/app11083321 fatcat:cyiwwkqiqvggdhb3zfqztbqqgi

LSTM-based Deep Learning Models for Non-factoid Answer Selection [article]

Ming Tan, Cicero dos Santos, Bing Xiang, Bowen Zhou
2016 arXiv   pre-print
In this paper, we apply a general deep learning (DL) framework for the answer selection task, which does not depend on manually defined features or linguistic tools.  ...  The basic framework is to build the embeddings of questions and answers based on bidirectional long short-term memory (biLSTM) models, and measure their closeness by cosine similarity.  ...  RESULTS CONCLUSION In this paper, we study the answer selection task by employing a bidirectional-LSTM based deep learning framework.  ... 
arXiv:1511.04108v4 fatcat:6hubuxrlxrampmjk6quic2st4i

SOLAR: Deep Structured Representations for Model-Based Reinforcement Learning [article]

Marvin Zhang, Sharad Vikram, Laura Smith, Pieter Abbeel, Matthew J. Johnson, Sergey Levine
2019 arXiv   pre-print
Model-based reinforcement learning (RL) has proven to be a data efficient approach for learning control tasks but is difficult to utilize in domains with complex observations such as images.  ...  In this paper, we present a method for learning representations that are suitable for iterative model-based policy improvement, even when the underlying dynamical system has complex dynamics and image  ...  For the real world tasks, we also compare to deep visual foresight (DVF; Ebert et al., 2018) , a state-of-the-art model-based method for images which does not use representation learning.  ... 
arXiv:1808.09105v4 fatcat:jpgdhn6b35ec5k3erhgdhp4ofy

Learning Tree-based Deep Model for Recommender Systems

Han Zhu, Xiang Li, Pengye Zhang, Guozheng Li, Jie He, Han Li, Kun Gai
2018 Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining - KDD '18  
We propose a novel tree-based method which can provide logarithmic complexity w.r.t. corpus size even with more expressive models such as deep neural networks.  ...  Model-based methods for recommender systems have been studied extensively in recent years.  ...  Then we propose the joint learning framework of the tree-based index and deep model. In the last subsection, we specify the hierarchical user preference representation used in model training.  ... 
doi:10.1145/3219819.3219826 dblp:conf/kdd/ZhuLZLHLG18 fatcat:zp7goqjacrgvbglzq2tv2gmapi

Auto-Ensemble: An Adaptive Learning Rate Scheduling based Deep Learning Model Ensembling [article]

Jun Yang, Fei Wang
2020 arXiv   pre-print
This paper proposes Auto-Ensemble (AE) to collect checkpoints of deep learning model and ensemble them automatically by adaptive learning rate scheduling algorithm.  ...  Ensembling deep learning models is a shortcut to promote its implementation in new scenarios, which can avoid tuning neural networks, losses and training algorithms from scratch.  ...  And the feature engineering is highly based on manual selection.To avoid huge training budget and complicated feature engineering, this paper attempts to provide a deep learning based simple and automatic  ... 
arXiv:2003.11266v1 fatcat:tc6rz5gl4jbdxebdcrsaqipylm

Algorithmic Framework for Model-based Deep Reinforcement Learning with Theoretical Guarantees [article]

Yuping Luo, Huazhe Xu, Yuanzhi Li, Yuandong Tian, Trevor Darrell, Tengyu Ma
2021 arXiv   pre-print
Model-based reinforcement learning (RL) is considered to be a promising approach to reduce the sample complexity that hinders model-free RL.  ...  This paper introduces a novel algorithmic framework for designing and analyzing model-based RL algorithms with theoretical guarantees.  ...  Despite promising empirical findings, many of theoretical properties of model-based deep reinforcement learning are not well-understood.  ... 
arXiv:1807.03858v5 fatcat:4g56r23yoje73g2imvc73nd7oy

Recent Progresses in Deep Learning based Acoustic Models (Updated) [article]

Dong Yu, Jinyu Li
2018 arXiv   pre-print
In this paper, we summarize recent progresses made in deep learning based acoustic models and the motivation and insights behind the surveyed techniques.  ...  , and the attention-based sequence-to-sequence model.  ...  With the application of deep learning models, now the ASR systems on close-talking scenario perform very well.  ... 
arXiv:1804.09298v2 fatcat:yfxzxu6qanbndcnmt3loikqeym
« Previous Showing results 1 — 15 out of 904,529 results