DOCSLIB.ORG

Text Simplification with Reinforcement Learning Using Supervised

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Text Simplification with Reinforcement Learning Using Supervised Text Simplification with Reinforcement Learning using Supervised Rewards on Grammaticality, Meaning Preservation, and Simplicity Akifumi Nakamachiy, Tomoyuki Kajiwaraz, Yuki Arasey yGraduate School of Information Science and Technology, Osaka University zInstitute for Datability Science, Osaka University yfnakamachi.akifumi, [email protected] [email protected] Abstract Complex Simple Sentence Encoder Decoder Sentence We optimize rewards of reinforcement learn- ing in text simplification using metrics that are highly correlated with human-perspectives. To address problems of exposure bias and loss- Grammaticality evaluation mismatch, text-to-text generation Meaning tasks employ reinforcement learning that re- Preservation wards task-specific metrics. Previous studies in text simplification employ the weighted sum Simplicity of sub-rewards from three perspectives: gram- Reward Calculator maticality, meaning preservation, and simplic- ity. However, the previous rewards do not Figure 1: Overview of the reinforcement learning for align with human-perspectives for these per- text simplification. spectives. In this study, we propose to use BERT regressors fine-tuned for grammatical- ity, meaning preservation, and simplicity as re- is evaluated as a whole sentence during inference, ward estimators to achieve text simplification it is evaluated at the token-level during training. conforming to human-perspectives. Experi- To address these problems, reinforcement learning mental results show that reinforcement learn- has been employed in text-to-text generation tasks, ing with our rewards balances meaning preser- vation and simplicity. Additionally, human such as machine translation (Ranzato et al., 2016) evaluation confirmed that simplified texts by and abstractive summarization (Paulus et al., 2018). our method are preferred by humans compared These studies use metrics suitable for each task, to previous studies. such as BLEU (Papineni et al., 2002) and ROUGE (Lin, 2004), as rewards. Although reinforcement 1 Introduction learning based text simplification models (Zhang Text simplification is one of the text-to-text gen- and Lapata, 2017; Zhao et al., 2020) have used re- eration tasks that rewrites complex sentences into wards metrics such as SARI (Xu et al., 2016) and simpler ones. Text simplification is useful for pre- FKGL (Kincaid et al., 1975), these metrics do not processing of NLP tasks such as semantic role la- align with human-perspectives, i.e., human evalu- beling (Vickrey and Koller, 2008; Woodsend and ation results (Xu et al., 2016; Sulem et al., 2018; Lapata, 2014) and machine translation (Stajnerˇ and Alva-Manchego et al., 2020). Popovic´, 2016, 2018). It also has valuable applica- In this study, we train a text simplification model tions such as assisting language learning (Inui et al., based on reinforcement learning with rewards that 2003; Petersen and Ostendorf, 2007) and helping highly agree with human-perspectives. Specifically, language-impaired readers (Carroll et al., 1999). we apply a BERT regressor (Devlin et al., 2019) There are two problems in text-to-text genera- on grammaticality, meaning preservation, and sim- tion with an encoder-decoder model: exposure bias plicity, respectively, as shown in Figure1. Exper- and loss-evaluation mismatch (Ranzato et al., 2016; iments on the Newsela dataset (Xu et al., 2015) Wiseman and Rush, 2016). The former is that the have shown that reinforcement learning with our model is not exposed to its own errors during train- rewards balances meaning preservation and sim- ing. The latter is that while the generated sentence plicity. Further, manual evaluation has shown that 153 Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing: Student Research Workshop, pages 153–159 December 4 - 7, 2020. c 2020 Association for Computational Linguistics our outputs were preferred by humans compared to As automatic evaluation metrics for text simpli- previous models. fication, BLEU, SARI, and FKGL have been used; however, there has not been a consensus of stan- 2 Background: Reinforcement Learning dard metrics because of their low correlation with for Text Simplification human perspectives (Xu et al., 2016; Sulem et al., 2018; Alva-Manchego et al., 2020). Therefore, the Reinforcement learning in text-to-text generation previous studies designed rewards from the follow- tasks is performed as additional training for pre- ing three perspectives, based on the standards in trained text-to-text generation models. It is a com- manual evaluation for text simplification. mon technique to linearly interpolate a reward of reinforcement learning and the cross-entropy loss • Grammaticality: This reward assesses the to avoid misleading training because of a large ac- grammatical acceptability of the generated tion space (Ranzato et al., 2016; Zhang and Lapata, sentence Y^ . Previous studies used an neu- 2017). We first explain an attention based encoder- ral language model implemented using Long decoder model (EncDecA) (Luong et al., 2015) in short-term memory (Mikolov et al., 2010; Section 2.1 and then reinforcement learning for text Hochreiter and Schmidhuber, 1997). simplification in Section 2.2. • Meaning Preservation: This reward assesses 2.1 Encoder-Decoder Model with Attention the semantic similarity between the source sentence X and the generated sentence Y^ . Let X = (x1; ··· ; xjXj) be a source sentence and Zhang and Lapata(2017) used cosine simi- Y = (y1; ··· ; yjY j) be its reference sentence. In text simplification, the source and reference are larity of the sentence representations from a complex and simple sentences, respectively. An sequence auto-encoder (Dai and Le, 2015). encoder takes a source sentence as input, and out- Zhao et al.(2020) used cosine similarity of puts hidden states. Decoder generates a word dis- sentence representations which consists of tribution at time step t + 1 from all the encoder weighted average of word embeddings (Arora hidden states and the series of decoder hidden state et al., 2017). ^ (h1; ··· ; ht). We generate a sentence Y by sam- • Simplicity: This reward assesses the sim- pling words from the distribution at each time step. plicity of the generated sentence Y^ . Zhang The objective function for training is averaged and Lapata(2017) used SARI (X; Y; Y^ ) score, cross entropy loss of sentence pairs: while Zhao et al.(2020) used FKGL (Y^ ) score. jY j X Among different ways to conduct reinforcement LC = − log P (yt+1jy1···t;X): (1) learning, one of the standard approaches used in t=1 text simplification is directly maximizing the re- wards by the REINFORCE algorithm (Williams, As the Equation (1) suggests, y1···t is given at train- ing but not at an inference (exposure bias situation). 1992; Ranzato et al., 2016). This approach opti- In addition, cross entropy loss cannot be evaluated mizes the log probability weighted by the expected at a sentence-level (loss-evaluation mismatch). future reward as the objective function: jY j 2.2 Reinforcement Learning X LR = − r(ht) log P (yt+1jy1···t;X); (2) Similar to other text-to-text generation tasks (Ran- t=1 zato et al., 2016; Paulus et al., 2018), reinforcement where the expected future reward r(h ) is estimated learning is applied for text simplification (Zhang t using a reward estimator R(·) and a baseline esti- and Lapata, 2017; Zhao et al., 2020) to address mator b(h ) calculated from the hidden state at time the problems of exposure bias and loss-evaluation t step t. mismatch. In the reinforcement learning step, the r(h ) = R(·) − b(h ): (3) pre-trained text-to-text generation model is trained t t to increase the reward R(·). By employing a re- Following (Ranzato et al., 2016), the baseline esti- 2 ward function that takes the entire sentence Y^ into mator is optimised by minimizing kbt − R(·)k . account, the exposure bias and loss-evaluation mis- Hashimoto and Tsuruoka(2019) discussed prob- match problems are mitigated. lems in text-to-text generation by reinforcement 154 learning; the expected future reward estimation Train Validation Test is unstable due to the huge action space, which GUG 1; 518 747 754 hinders convergence. This is because the action STS-B 5; 749 1; 500 1; 379 space of text-to-text generation corresponds to the Newsela 94; 208 1; 129 1; 077 entire target vocabulary, where many words are rarely used for prediction. Therefore, previous stud- Table 1: The numbers of sentences in datasets for each ies (Wu et al., 2018; Paulus et al., 2018; Hashimoto sub-reward estimator and Tsuruoka, 2019) proposed to stabilize the train- ing in reinforcement learning by first pre-training a model with cross-entropy loss, and then adding of a sentence. The GUG dataset consists of sen- weighted REINFORCE loss: tences written by English as the second language learners. Each sentence has four native English L = λLR + (1 − λ)LC : (4) speakers assessing grammatical acceptability on a scale of 1 to 4. We estimate the average of these 3 BERT-based Supervised Reward ratings. We propose a reward estimator R(X; Y^ ) consisting Meaning Preservation We use the STS-B 3 of sub-rewards for grammaticality RG, meaning dataset (Cer et al., 2017) for estimating the mean- preservation RM, and simplicity RS. These sub- ing preservation of sentence pairs. The STS-B rewards are combined by weighted sum with hyper dataset consists of sentence pairs from multiple parameters of δ and : sources such as news headlines and image captions. Each sentence pair is evaluated for semantic sim- ^ ^ ^ ilarity by five cloud workers on a scale of 0 to 5. R(X; Y ) = δRG(Y ) + RM(X; Y ) (5) We estimate the average of these ratings. + (1 − δ − )RS(Y^ ): Simplicity We use the Newsela dataset4 (Xu To achieve a better correlation between each sub- et al., 2015) for estimating the simplicity of a sen- reward and human perspectives, we employ BERT tence. The Newsela dataset is a parallel dataset of regressors and fine-tune them using manually an- complex and simple sentences.
Load more

Recommended publications
  • Survey on Reinforcement Learning for Language Processing
    Survey on reinforcement learning for language processing V´ıctorUc-Cetina1, Nicol´asNavarro-Guerrero2, Anabel Martin-Gonzalez1, Cornelius Weber3, Stefan Wermter3 1 Universidad Aut´onomade Yucat´an- fuccetina, [email protected] 2 Aarhus University - [email protected] 3 Universit¨atHamburg - fweber, [email protected] February 2021 Abstract In recent years some researchers have explored the use of reinforcement learning (RL) algorithms as key components in the solution of various natural language process- ing tasks. For instance, some of these algorithms leveraging deep neural learning have found their way into conversational systems. This paper reviews the state of the art of RL methods for their possible use for different problems of natural language processing, focusing primarily on conversational systems, mainly due to their growing relevance. We provide detailed descriptions of the problems as well as discussions of why RL is well-suited to solve them. Also, we analyze the advantages and limitations of these methods. Finally, we elaborate on promising research directions in natural language processing that might benefit from reinforcement learning. Keywords| reinforcement learning, natural language processing, conversational systems, pars- ing, translation, text generation 1 Introduction Machine learning algorithms have been very successful to solve problems in the natural language pro- arXiv:2104.05565v1 [cs.CL] 12 Apr 2021 cessing (NLP) domain for many years, especially supervised and unsupervised methods. However, this is not the case with reinforcement learning (RL), which is somewhat surprising since in other domains, reinforcement learning methods have experienced an increased level of success with some impressive results, for instance in board games such as AlphaGo Zero [106].
    [Show full text]
  • A Deep Reinforcement Learning Neural Network Folding Proteins
    DeepFoldit - A Deep Reinforcement Learning Neural Network Folding Proteins Dimitra Panou1, Martin Reczko2 1University of Athens, Department of Informatics and Telecommunications 2Biomedical Sciences Research Center “Alexander Fleming” ABSTRACT Despite considerable progress, ab initio protein structure prediction remains suboptimal. A crowdsourcing approach is the online puzzle video game Foldit [1], that provided several useful results that matched or even outperformed algorithmically computed solutions [2]. Using Foldit, the WeFold [3] crowd had several successful participations in the Critical Assessment of Techniques for Protein Structure Prediction. Based on the recent Foldit standalone version [4], we trained a deep reinforcement neural network called DeepFoldit to improve the score assigned to an unfolded protein, using the Q-learning method [5] with experience replay. This paper is focused on model improvement through hyperparameter tuning. We examined various implementations by examining different model architectures and changing hyperparameter values to improve the accuracy of the model. The new model’s hyper-parameters also improved its ability to generalize. Initial results, from the latest implementation, show that given a set of small unfolded training proteins, DeepFoldit learns action sequences that improve the score both on the training set and on novel test proteins. Our approach combines the intuitive user interface of Foldit with the efficiency of deep reinforcement learning. KEYWORDS: ab initio protein structure prediction, Reinforcement Learning, Deep Learning, Convolution Neural Networks, Q-learning 1. ALGORITHMIC BACKGROUND Machine learning (ML) is the study of algorithms and statistical models used by computer systems to accomplish a given task without using explicit guidelines, relying on inferences derived from patterns. ML is a field of artificial intelligence.
    [Show full text]
  • An Introduction to Deep Reinforcement Learning
    An Introduction to Deep Reinforcement Learning Ehsan Abbasnejad Remember: Supervised Learning We have a set of sample observations, with labels learn to predict the labels, given a new sample cat Learn the function that associates a picture of a dog/cat with the label dog Remember: supervised learning We need thousands of samples Samples have to be provided by experts There are applications where • We can’t provide expert samples • Expert examples are not what we mimic • There is an interaction with the world Deep Reinforcement Learning AlphaGo Scenario of Reinforcement Learning Observation Action State Change the environment Agent Don’t do that Reward Environment Agent learns to take actions maximizing expected Scenario of Reinforcement Learningreward. Observation Action State Change the environment Agent Thank you. Reward https://yoast.com/how-t Environment o-clean-site-structure/ Machine Learning Actor/Policy ≈ Looking for a Function Action = π( Observation ) Observation Action Function Function input output Used to pick the Reward best function Environment Reinforcement Learning in a nutshell RL is a general-purpose framework for decision-making • RL is for an agent with the capacity to act • Each action influences the agent’s future state • Success is measured by a scalar reward signal Goal: select actions to maximise future reward Deep Learning in a nutshell DL is a general-purpose framework for representation learning • Given an objective • Learning representation that is required to achieve objective • Directly from raw inputs
    [Show full text]
  • Modelling Working Memory Using Deep Recurrent Reinforcement Learning
    Modelling Working Memory using Deep Recurrent Reinforcement Learning Pravish Sainath1;2;3 Pierre Bellec1;3 Guillaume Lajoie 2;3 1Centre de Recherche, Institut Universitaire de Gériatrie de Montréal 2Mila - Quebec AI Institute 3Université de Montréal Abstract 1 In cognitive systems, the role of a working memory is crucial for visual reasoning 2 and decision making. Tremendous progress has been made in understanding the 3 mechanisms of the human/animal working memory, as well as in formulating 4 different frameworks of artificial neural networks. In the case of humans, the [1] 5 visual working memory (VWM) task is a standard one in which the subjects are 6 presented with a sequence of images, each of which needs to be identified as to 7 whether it was already seen or not. Our work is a study of multiple ways to learn a 8 working memory model using recurrent neural networks that learn to remember 9 input images across timesteps in supervised and reinforcement learning settings. 10 The reinforcement learning setting is inspired by the popular view in Neuroscience 11 that the working memory in the prefrontal cortex is modulated by a dopaminergic 12 mechanism. We consider the VWM task as an environment that rewards the 13 agent when it remembers past information and penalizes it for forgetting. We 14 quantitatively estimate the performance of these models on sequences of images [2] 15 from a standard image dataset (CIFAR-100 ) and their ability to remember 16 and recall. Based on our analysis, we establish that a gated recurrent neural 17 network model with long short-term memory units trained using reinforcement 18 learning is powerful and more efficient in temporally consolidating the input spatial 19 information.
    [Show full text]
  • Reinforcement Learning: a Survey
    Journal of Articial Intelligence Research Submitted published Reinforcement Learning A Survey Leslie Pack Kaelbling lpkcsbrownedu Michael L Littman mlittmancsbrownedu Computer Science Department Box Brown University Providence RI USA Andrew W Mo ore awmcscmuedu Smith Hal l Carnegie Mel lon University Forbes Avenue Pittsburgh PA USA Abstract This pap er surveys the eld of reinforcement learning from a computerscience p er sp ective It is written to b e accessible to researchers familiar with machine learning Both the historical basis of the eld and a broad selection of current work are summarized Reinforcement learning is the problem faced by an agent that learns b ehavior through trialanderror interactions with a dynamic environment The work describ ed here has a resemblance to work in psychology but diers considerably in the details and in the use of the word reinforcement The pap er discusses central issues of reinforcement learning including trading o exploration and exploitation establishing the foundations of the eld via Markov decision theory learning from delayed reinforcement constructing empirical mo dels to accelerate learning making use of generalization and hierarchy and coping with hidden state It concludes with a survey of some implemented systems and an assessment of the practical utility of current metho ds for reinforcement learning Intro duction Reinforcement learning dates back to the early days of cyb ernetics and work in statistics psychology neuroscience and computer science In the last ve to ten years
    [Show full text]
  • An Energy Efficient Edgeai Autoencoder Accelerator For
    Received 26 July 2020; revised 29 October 2020; accepted 28 November 2020. Date of current version 26 January 2021. Digital Object Identifier 10.1109/OJCAS.2020.3043737 An Energy Efficient EdgeAI Autoencoder Accelerator for Reinforcement Learning NITHEESH KUMAR MANJUNATH1 (Member, IEEE), AIDIN SHIRI1, MORTEZA HOSSEINI1 (Graduate Student Member, IEEE), BHARAT PRAKASH1, NICHOLAS R. WAYTOWICH2, AND TINOOSH MOHSENIN1 1Department of Computer Science and Electrical Engineering, University of Maryland, Baltimore, MD 21250, USA 2U.S. Army Research Laboratory, Aberdeen, MD, USA This article was recommended by Associate Editor Y. Li. CORRESPONDING AUTHOR: N. K. MANJUNATH (e-mail: [email protected]) This work was supported by the U.S. Army Research Laboratory through Cooperative Agreement under Grant W911NF-10-2-0022. ABSTRACT In EdgeAI embedded devices that exploit reinforcement learning (RL), it is essential to reduce the number of actions taken by the agent in the real world and minimize the compute-intensive policies learning process. Convolutional autoencoders (AEs) has demonstrated great improvement for speeding up the policy learning time when attached to the RL agent, by compressing the high dimensional input data into a small latent representation for feeding the RL agent. Despite reducing the policy learning time, AE adds a significant computational and memory complexity to the model which contributes to the increase in the total computation and the model size. In this article, we propose a model for speeding up the policy learning process of RL agent with the use of AE neural networks, which engages binary and ternary precision to address the high complexity overhead without deteriorating the policy that an RL agent learns.
    [Show full text]
  • Sparse Cooperative Q-Learning
    Sparse Cooperative Q-learning Jelle R. Kok [email protected] Nikos Vlassis [email protected] Informatics Institute, Faculty of Science, University of Amsterdam, The Netherlands Abstract in uncertain environments. In principle, it is pos- sible to treat a multiagent system as a `big' single Learning in multiagent systems suffers from agent and learn the optimal joint policy using stan- the fact that both the state and the action dard single-agent reinforcement learning techniques. space scale exponentially with the number of However, both the state and action space scale ex- agents. In this paper we are interested in ponentially with the number of agents, rendering this using Q-learning to learn the coordinated ac- approach infeasible for most problems. Alternatively, tions of a group of cooperative agents, us- we can let each agent learn its policy independently ing a sparse representation of the joint state- of the other agents, but then the transition model de- action space of the agents. We first examine pends on the policy of the other learning agents, which a compact representation in which the agents may result in oscillatory behavior. need to explicitly coordinate their actions only in a predefined set of states. Next, we On the other hand, in many problems the agents only use a coordination-graph approach in which need to coordinate their actions in few states (e.g., two we represent the Q-values by value rules that cleaning robots that want to clean the same room), specify the coordination dependencies of the while in the rest of the states the agents can act in- agents at particular states.
    [Show full text]
  • 4 Perceptron Learning
    4 Perceptron Learning 4.1 Learning algorithms for neural networks In the two preceding chapters we discussed two closely related models, McCulloch–Pitts units and perceptrons, but the question of how to find the parameters adequate for a given task was left open. If two sets of points have to be separated linearly with a perceptron, adequate weights for the comput- ing unit must be found. The operators that we used in the preceding chapter, for example for edge detection, used hand customized weights. Now we would like to find those parameters automatically. The perceptron learning algorithm deals with this problem. A learning algorithm is an adaptive method by which a network of com- puting units self-organizes to implement the desired behavior. This is done in some learning algorithms by presenting some examples of the desired input- output mapping to the network. A correction step is executed iteratively until the network learns to produce the desired response. The learning algorithm is a closed loop of presentation of examples and of corrections to the network parameters, as shown in Figure 4.1. network test input-output compute the examples error fix network parameters Fig. 4.1. Learning process in a parametric system R. Rojas: Neural Networks, Springer-Verlag, Berlin, 1996 78 4 Perceptron Learning In some simple cases the weights for the computing units can be found through a sequential test of stochastically generated numerical combinations. However, such algorithms which look blindly for a solution do not qualify as “learning”. A learning algorithm must adapt the network parameters accord- ing to previous experience until a solution is found, if it exists.
    [Show full text]
  • Optimal Path Routing Using Reinforcement Learning
    OPTIMAL PATH ROUTING USING REINFORCEMENT LEARNING Rajasekhar Nannapaneni Sr Principal Engineer, Solutions Architect Dell EMC [email protected] Knowledge Sharing Article © 2020 Dell Inc. or its subsidiaries. The Dell Technologies Proven Professional Certification program validates a wide range of skills and competencies across multiple technologies and products. From Associate, entry-level courses to Expert-level, experience-based exams, all professionals in or looking to begin a career in IT benefit from industry-leading training and certification paths from one of the world’s most trusted technology partners. Proven Professional certifications include: • Cloud • Converged/Hyperconverged Infrastructure • Data Protection • Data Science • Networking • Security • Servers • Storage • Enterprise Architect Courses are offered to meet different learning styles and schedules, including self-paced On Demand, remote-based Virtual Instructor-Led and in-person Classrooms. Whether you are an experienced IT professional or just getting started, Dell Technologies Proven Professional certifications are designed to clearly signal proficiency to colleagues and employers. Learn more at www.dell.com/certification 2020 Dell Technologies Proven Professional Knowledge Sharing 2 Abstract Optimal path management is key when applied to disk I/O or network I/O. The efficiency of a storage or a network system depends on optimal routing of I/O. To obtain optimal path for an I/O between source and target nodes, an effective path finding mechanism among a set of given nodes is desired. In this article, a novel optimal path routing algorithm is developed using reinforcement learning techniques from AI. Reinforcement learning considers the given topology of nodes as the environment and leverages the given latency or distance between the nodes to determine the shortest path.
    [Show full text]
  • 18 Reinforcement Learning
    18 Reinforcement Learning In reinforcement learning, the learner is a decision-making agent that takes actions in an environment and receives reward (or penalty) for its actions in trying to solve a problem. After a set of trial-and- error runs, it should learn the best policy, which is the sequence of actions that maximize the total reward. 18.1 Introduction Let us say we want to build a machine that learns to play chess. In this case we cannot use a supervised learner for two reasons. First, it is very costly to have a teacher that will take us through many games and indicate us the best move for each position. Second, in many cases, there is no such thing as the best move; the goodness of a move depends on the moves that follow. A single move does not count; a sequence of moves is good if after playing them we win the game. The only feedback is at the end of the game when we win or lose the game. Another example is a robot that is placed in a maze. The robot can move in one of the four compass directions and should make a sequence of movements to reach the exit. As long as the robot is in the maze, there is no feedback and the robot tries many moves until it reaches the exit and only then does it get a reward. In this case there is no opponent, but we can have a preference for shorter trajectories, implying that in this case we play against time.
    [Show full text]
  • Deep Learning and Reinforcement Learning
    Deep learning and reinforcement learning Jes´usFern´andez-Villaverde1 and Galo Nu~no2 June 27, 2021 1University of Pennsylvania 2Banco de Espa~na Course outline 1. Dynamic programming in continuous time. 2. Deep learning and reinforcement learning. 3. Heterogeneous agent models. 4. Optimal policies with heterogeneous agents. 1 A short introduction The problem • Let us supposed we want to approximate an unknown function: y = f (x) where y is a scalar and x = x1; x2; :::; xN a vector. f g • Easy to generalize to the case where y is a vector (or a probability distribution), but notation becomes cumbersome. • In economics, f (x) can be a value function, a policy function, a pricing kernel, a conditional expectation, a classifier, ... 2 A neural network • An artificial neural network (ANN or connectionist system) is an approximation to f (x) built as a linear combination of M generalized linear models of x of the form: M NN X y ∼= g (x; θ) = θ0 + θmφ (zm) m=1 where φ( ) is an arbitrary activation function and: · N X zm = θ0;m + θn;mxn n=1 • M is known as the width of the model. • We can select θ such that g NN (x; θ) is as close to f (x) as possible given some relevant metric. • This is known as \training" the network. 3 Comparison with other approximations • Compare: M N ! NN X X y ∼= g (x; θ) = θ0 + θmφ θ0;m + θn;mxn m=1 n=1 with a standard projection: M CP X y ∼= g (x; θ) = θ0 + θmφm (x) m=1 where φm is, for example, a Chebyshev polynomial.
    [Show full text]
  • Deep Reinforcement Learning for Imbalanced Classification
    Deep Reinforcement Learning for Imbalanced Classification Enlu Lin1 Qiong Chen2,* Xiaoming Qi3 School of Computer Science and Engineering South China University of Technology Guangzhou, China [email protected] [email protected] [email protected] Abstract—Data in real-world application often exhibit skewed misclassification cost to the minority class. However, with the class distribution which poses an intense challenge for machine rapid developments of big data, a large amount of complex learning. Conventional classification algorithms are not effective data with high imbalanced ratio is generated which brings an in the case of imbalanced data distribution, and may fail when the data distribution is highly imbalanced. To address this issue, we enormous challenge in imbalanced data classification. Con- propose a general imbalanced classification model based on deep ventional methods are inadequate to cope with more and reinforcement learning. We formulate the classification problem more complex data so that novel deep learning approaches as a sequential decision-making process and solve it by deep Q- are increasingly popular. learning network. The agent performs a classification action on In recent years, deep reinforcement learning has been one sample at each time step, and the environment evaluates the classification action and returns a reward to the agent. The successfully applied to computer games, robots controlling, reward from minority class sample is larger so the agent is more recommendation systems [5]–[7] and so on. For classification sensitive to the minority class. The agent finally finds an optimal problems, deep reinforcement learning has served in eliminat- classification policy in imbalanced data under the guidance of ing noisy data and learning better features, which made a great specific reward function and beneficial learning environment.
    [Show full text]