Model-Free Reinforcement Learning Algorithms: A Survey

Calisir, Sinan; Pehlıvanoğlu, Meltem Kurt

doi:10.1109/siu.2019.8806389

Cited by 17 publications

(10 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…RL studies the way natural and artificial systems can learn to predict the consequences of and optimize their behavior in environments where actions lead them from one state or situation to the next and can lead to rewards and punishments. A common model for RL is the standard Markov Decision Process. , RL can be divided into model-based RL and model-free RL, , as well as active RL and passive RL . DL models can also be used in RL to form deep RL (DRL). , These methods have been applied to diverse fields in biology and chemistry, including drug discovery, , protein design, , and chemical engineering. , They have also been used in image recognition , and financial markets .…”

Section: Methods For Small Molecular Data Challengesmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning Methods for Small Data Challenges in Molecular Science

et al. 2023

View full text Add to dashboard Cite

Small data are often used in scientific and engineering research due to the presence of various constraints, such as time, cost, ethics, privacy, security, and technical limitations in data acquisition. However, big data have been the focus for the past decade, small data and their challenges have received little attention, even though they are technically more severe in machine learning (ML) and deep learning (DL) studies. Overall, the small data challenge is often compounded by issues, such as data diversity, imputation, noise, imbalance, and high-dimensionality. Fortunately, the current big data era is characterized by technological breakthroughs in ML, DL, and artificial intelligence (AI), which enable data-driven scientific discovery, and many advanced ML and DL technologies developed for big data have inadvertently provided solutions for small data problems. As a result, significant progress has been made in ML and DL for small data challenges in the past decade. In this review, we summarize and analyze several emerging potential solutions to small data challenges in molecular science, including chemical and biological sciences. We review both basic machine learning algorithms, such as linear regression, logistic regression (LR), k-nearest neighbor (KNN), support vector machine (SVM), kernel learning (KL), random forest (RF), and gradient boosting trees (GBT), and more advanced techniques, including artificial neural network (ANN), convolutional neural network (CNN), U-Net, graph neural network (GNN), Generative Adversarial Network (GAN), long short-term memory (LSTM), autoencoder, transformer, transfer learning, active learning, graph-based semi-supervised learning, combining deep learning with traditional machine learning, and physical model-based data augmentation. We also briefly discuss the latest advances in these methods. Finally, we conclude the survey with a discussion of promising trends in small data challenges in molecular science.

show abstract

Section: Methods For Small Molecular Data Challengesmentioning

confidence: 99%

“…A common model for RL is the standard Markov Decision Process. 354,355 RL can be divided into model-based RL 356 and model-free RL, 357,358 as well as active RL 359 and passive RL. 360 DL models can also be used in RL to form deep RL (DRL).…”

Section: Reinforcement Learningmentioning

confidence: 99%

Machine Learning Methods for Small Data Challenges in Molecular Science

et al. 2023

View full text Add to dashboard Cite

show abstract

“…They directly learn from experience by estimating the optimal policy or value function. Model-free algorithms such as Q-learning, Monte Carlo Control, SARSA (State-Action-Reward-State-Action), Deep Q Network, etc., are widely used in RL [24].…”

Section: Reinforcement Learningmentioning

confidence: 99%

Application of Artificial Intelligence in Biochemistry Research: A Review

Jada,

Stephen,

Usman

et al. 2023

J Biochem Microbiol Biotechnol

View full text Add to dashboard Cite

The exploration of Biochemistry research has been expanded by artificial intelligence (AI) and its ability to analyse immense and intricate datasets in a way that would be unattainable by human effort alone. This review delves into the most recent examples of AI breakthroughs that had a transformative impact on key aspects of biochemistry. AI has now led to the creation and improvement of drug molecules and the capability to predict which new proteins could be targeted for repurposing with current drugs. When it comes to protein structures, algorithms such as AlphaFold have made great strides in resolving the protein folding problem that has been a challenge for so long. Reliably identifying proteins and metabolites from spectral data is now possible with deep learning models. Meanwhile, AI can classify sequences and spot gene expression patterns in massive genomics and transcriptomics datasets with ease. The remarkable capabilities of AI to automate the analysis of medical images and natural language descriptions of patient symptoms have a promising potential for transforming disease diagnosis and treatment. Nevertheless, obstacles such as data availability, interpretability of AI models, ethical considerations, and generalization must be tackled as these technologies evolve. The collaboration between AI and biochemistry appears to be optimistic, with biochemical data powering the development of more robust AI systems that can extract new knowledge from vast datasets beyond human reach. Thus, this mutually beneficial relationship has the potential to vastly expedite discovery across molecular biology.

show abstract

“…But, it needs to improve the routing efficiency. A secure and efficient privacy-preserving unspecified authentication approach offers data security and privacy [18]. A trust-based formal model depicts the fault identification procedure and confirms faults [19].…”

Section: Introductionmentioning

confidence: 99%

Bayesian decision model based reliable route formation in internet of things

Kumar,

Mishra,

Reddy

et al. 2024

IJEECS

View full text Add to dashboard Cite

Security provisioning has become an important issue in wireless multimedia networks because of their crucial task of supporting several services. This paper presents Bayesian decision model based reliable route formation in internet of things (BDMI). The main objective of the BDMI approach is to distinguish unreliable sensor nodes and transmit the data efficiently. Active and passive attack recognition methods identify unreliable node sensor nodes. Remaining energy, node degree, and packet transmission rate parameters to observe their node possibilities for recognizing the passive unreliable nodes. In active recognition, the base station (BS) confirms every sensor node identity, remaining energy, supportive node rate, node location, and link efficiency parameters to detect active unreliable sensor nodes. The Bayesian decision model (BDM) efficiently isolates an unreliable sensor node in the multimedia network. Simulation outcomes illustrate that the BDMI approach can efficiently enhance unreliable node detection and minimize the packet loss ratio in the network.

show abstract

Model-Free Reinforcement Learning Algorithms: A Survey

Cited by 17 publications

References 5 publications

Machine Learning Methods for Small Data Challenges in Molecular Science

Machine Learning Methods for Small Data Challenges in Molecular Science

Application of Artificial Intelligence in Biochemistry Research: A Review

Bayesian decision model based reliable route formation in internet of things

Contact Info

Product

Resources

About