scholarly journals Neural mechanisms of distributed value representations and learning strategies

2021 ◽  
Author(s):  
Shiva Farashahi ◽  
Alireza Soltani
Keyword(s):  
Neural Networks ◽  
Learning Strategies ◽  
Recurrent Neural Networks ◽  
Learning Strategy ◽  
Learning Task ◽  
Neural Mechanisms ◽  
Probabilistic Learning ◽  
Informative Feature ◽  
Alternative Solutions ◽  
Combining Estimates

AbstractLearning appropriate representations of the reward environment is extremely challenging in the real world where there are many options to learn about and these options have many attributes or features. Despite existence of alternative solutions for this challenge, neural mechanisms underlying emergence and adoption of value representations and learning strategies remain unknown. To address this, we measured learning and choice during a novel multi-dimensional probabilistic learning task in humans and trained recurrent neural networks (RNNs) to capture our experimental observations. We found that participants estimate stimulus-outcome associations by learning and combining estimates of reward probabilities associated with the informative feature followed by those of informative conjunctions. Through analyzing representations, connectivity, and lesioning of the RNNs, we demonstrate this mixed learning strategy relies on a distributed neural code and distinct contributions of inhibitory and excitatory neurons. Together, our results reveal neural mechanisms underlying emergence of complex learning strategies in naturalistic settings.

scholarly journals Computational mechanisms of distributed value representations and mixed learning strategies

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Shiva Farashahi ◽  
Alireza Soltani
Keyword(s):  
Learning Strategies ◽  
Learning Strategy ◽  
Learning Task ◽  
Neural Mechanisms ◽  
Inhibitory Neurons ◽  
Probabilistic Learning ◽  
Informative Feature ◽  
Human Participants ◽  
Alternative Solutions ◽  
Combining Estimates

AbstractLearning appropriate representations of the reward environment is challenging in the real world where there are many options, each with multiple attributes or features. Despite existence of alternative solutions for this challenge, neural mechanisms underlying emergence and adoption of value representations and learning strategies remain unknown. To address this, we measure learning and choice during a multi-dimensional probabilistic learning task in humans and trained recurrent neural networks (RNNs) to capture our experimental observations. We find that human participants estimate stimulus-outcome associations by learning and combining estimates of reward probabilities associated with the informative feature followed by those of informative conjunctions. Through analyzing representations, connectivity, and lesioning of the RNNs, we demonstrate this mixed learning strategy relies on a distributed neural code and opponency between excitatory and inhibitory neurons through value-dependent disinhibition. Together, our results suggest computational and neural mechanisms underlying emergence of complex learning strategies in naturalistic settings.

scholarly journals ELT Teachers Attitude towards Language Learning Strategies Used in Southern Punjab, Pakistan

2019 ◽  
Vol IV (I) ◽  
pp. 271-280
Author(s):  
Abdul Khaliq ◽  
Akbar Ali ◽  
Fazal Hanan
Keyword(s):  
Language Learning ◽  
Learning Strategies ◽  
Learning Process ◽  
Teaching Methods ◽  
Learning Strategy ◽  
Learning Task ◽  
Language Learning Strategies ◽  
Learning And Teaching ◽  
Learning Techniques ◽  
Almost All

The present study throws light on language learning strategies, their effect on learning and instructors attitude in this respect .It defines that a learning strategy is a learners approach of understanding and employing particular skills in order to accomplish learning task efficiently. It also stresses that todays learner is smart enough to devise ways and methods to accelerate learning process. Learners use these techniques according to their needs and stage of learning. In parallel, it also explains that these techniques effect the behavior of instructor and his teaching methods as well. The researcher collected data from 110 participants of different schools, colleges and universities of Dera Ghazi (DG) Khan through questionnaire. This study shows that almost all the learners andteachers are inclined to use different learning techniques and improve their performance in this way. The researcher also identified strategies that are commonly used by learners and teachers to facilitate learning and teaching.

scholarly journals Recursive Processing of Nested Structures in Monkeys? Two alternative accounts

2021 ◽  
Author(s):  
Yair Lakretz ◽  
Stanislas Dehaene
Keyword(s):  
Neural Networks ◽  
Simple Model ◽  
Recurrent Neural Networks ◽  
Neural Mechanisms ◽  
Macaque Monkeys ◽  
Additional Training ◽  
Embedded Structures ◽  
Nested Structures ◽  
Sequence Memory ◽  
Additional Interpretation

Ferrigno et al. [2020] introduced an ingenious task to investigate recursion in human and non-human primates. American adults, Tsimane adults, and 3-5 year-old children successfully performed the task. Macaque monkeys required additional training, but two out of three eventually showed good generalization and scored above many Tsimane and child participants. Moreover, when tested on sequences composed of new bracket signs, the monkeys still showed good performance. The authors thus concluded that recursive nesting is not unique to humans. Here, we dispute the claim by showing that at least two alternative interpretations remain tenable. We first examine this conclusion in light of recent findings from modern artificial recurrent neural networks (RNNs), regarding how these networks encode sequences. We show that although RNNs, like monkeys, succeed on demanding generalization tasks as in Ferrigno et al., the underlying neural mechanisms are not recursive. Moreover, we show that when the networks are tested on sequences with deeper center-embedded structures compared to training, the networks fail to generalize. We then discuss an additional interpretation of the results in light of a simple model of sequence memory.

DISTRIBUTED LEARNING STRATEGY BASED ON CHIPS FOR CLASSIFICATION WITH LARGE-SCALE DATASET

2007 ◽  
Vol 21 (05) ◽  
pp. 899-920 ◽  
Author(s):  
BO YANG ◽  
XIAOHONG SU ◽  
YADONG WANG
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Secondary Structure ◽  
Learning Strategies ◽  
Large Scale ◽  
Learning Strategy ◽  
Protein Secondary Structure ◽  
Distributed Learning ◽  
Large Scale Dataset ◽  
Artificial Neural

Learning with very large-scale datasets is always necessary when handling real problems using artificial neural networks. However, it is still an open question how to balance computing efficiency and learning stability, when traditional neural networks spend a large amount of running time and memory to solve a problem with large-scale learning dataset. In this paper, we report the first evaluation of neural network distributed-learning strategies in large-scale classification over protein secondary structure. Our accomplishments include: (1) an architecture analysis on distributed-learning, (2) the development of scalable distributed system for large-scale dataset classification, (3) the description of a novel distributed-learning strategy based on chips, (4) a theoretical analysis of distributed-learning strategies for structure-distributed and data-distributed, (5) an investigation and experimental evaluation of distributed-learning strategy based-on chips with respect to time complexity and their effect on the classification accuracy of artificial neural networks. It is demonstrated that the novel distributed-learning strategy is better-balanced in parallel computing efficiency and stability as compared with the previous algorithms. The application of the protein secondary structure prediction demonstrates that this method is feasible and effective in practical applications.

STRATEGI PEMBELAJARAN DAN MOTIF BERPRESTASI TERHADAP HASIL BELAJAR MATA PELAJARAN MENGGAMBAR TEKNIK

2013 ◽  
Vol 6 (1) ◽  
Author(s):  
Timbul Purba ◽  
Harun Sitompul
Keyword(s):  
Student Learning ◽  
Learning Outcomes ◽  
Learning Strategies ◽  
Learning Strategy ◽  
Student Learning Outcomes ◽  
High Achievement ◽  
Low Achievement ◽  
Significance Level ◽  
Achievement Motive ◽  
Analysis Technique

Abstrak: Penelitian ini bertujuan: (1) hasil belajar menggambar teknik siswa yang diajar dengan strategi pembelajaran elaborasi lebih tinggi dibandingkan dengan siswa yang diajar dengan strategi pembelajaran ekspositori, (2) hasil belajar menggambar teknik siswa yang memiliki motif berprestasi tinggi lebih tinggi dibandingkan dengan siswa yang memiliki motif berprestasi rendah dan (3) interaksi antara strategi pembelajaran dengan motif berprestasi dalam mempengaruhi hasil belajar menggambar teknik siswa. Metode penelitian menggunakan metode quasi eksperimen dengan desain penelitian faktorial 2x2, sedangkan teknik analisis data menggunakan ANAVA dua jalur pada taraf signifikansi a = 0.05. Hasil penelitian diperoleh: (1) hasil belajar menggambar teknik siswa yang diajar dengan strategi pembelajaran elaborasi lebih tinggi dibandingkan dengan hasil belajar siswa yang diajar dengan strategi pembelajaran ekspositori, (2) hasil belajar menggambar teknik siswa yang memiliki motif berprestasi tinggi lebih tinggi dibandingkan dengan hasil belajar siswa yang memiliki motif berprestasi rendah dan (3) terdapat interaksi antara strategi pembelajaran dengan motif berprestasi dalam mempengaruhi hasil belajar menggambar teknik siswa.   Kata Kunci: strategi pembelajaran elaborasi dan ekspositori, motif berprestasi, hasil belajar menggambar teknik   Abstract: This research was aimed to: (1) the learning outcomes of students who are taught drawing techniques with learning strategy elaboration higher than students taught by expository learning strategy, (2) drawing techniques learning outcomes of students who have high achievement motive higher than students who have low achievement motive, and (3) the interaction between learning strategy and achievement motives in affecting student learning outcomes drawing techniques. The research method used was quasi experiment with 2 x 2 factorial design. The analysis technique used is the two-track analysis of variance ANOVA (2 x 2) with a significance level α = 0.05. The findings of the study indicate: (1) the learning outcomes of students who are taught drawing techniques with learning strategy elaboration higher learning outcomes than students taught by expository learning strategy; (2) drawing techniques learning outcomes of students who have high achievement motive higher than the learning outcomes of students who have low achievement motive; and (3) there is interaction between learning strategy and achievement motives in affecting student learning outcomes drawing techniques. Keywords: elaboration learning strategies and expository, achievement motive, the result of learning drawing techniques

PENGARUH STRATEGI PEMBELAJARAN DAN TIPE KEPRIBADIAN SISWA TERHADAP HASIL BELAJAR EKONOMI

2017 ◽  
Vol 10 (2) ◽  
pp. 151
Author(s):  
Harningsih Fitri Situmorang
Keyword(s):  
Learning Outcomes ◽  
Learning Strategies ◽  
Learning Strategy ◽  
Problem Based Learning ◽  
Personality Type ◽  
Student Learning Outcomes ◽  
Personality Types ◽  
Descriptive Statistics ◽  
Scheffe Test ◽  
Inferential Statistic

Abstrak: Penelitian ini bertujuan :(1) Untuk mengetahui hasil belajar ekonomi siswa yang diajar dengan strategi pembelajaran berbasis masalah lebih tinggi dari siswa yang diajar dengan strategi pembelajaran ekspositori. (2) Untuk mengetahui hasil belajar  ekonomi siswa yang memiliki tipe kepribadian ekstrovert dan siswa yang memiliki kepribadian introvert. (3) Untuk mengetahui interaksi antara strategi pembelajaran dengan tipe kepribadian  terhadap hasil belajar Ekonomi. Metode penelitian yang digunakan adalah kuasi eksperimen dengan desain faktorial 2 x 2. Uji statistik yang digunakan adalah statistik deskriptif untuk menyajikan data dan dilanjutkan dengan statistik inferensial dengan menggunakan ANAVA dua jalur dengan taraf signifikan α = 0,05 yang dilanjutkan dengan uji Scheffe. Hasil penelitian menunjukkan: (1) hasil belajar ekonomi siswa yang diajarkan dengan strategi pembelajaran berbasis masalah lebih tinggi dari pada hasil belajar ekonomi siswa yang diajarkan dengan strategi pembelajaran ekspositori; (2) hasil belajar ekonomi siswa yang memiliki kepribadian ekstrovert lebih tinggi dari pada hasil belajar ekonomi siswa yang memiliki tipe kepribadian introvert; (3) terdapat interaksi antara strategi pembelajaran dengan tipe kepribadian  dalam mempengaruhi hasil belajar siswa. Hipotesis ini menunjukkan bahwa strategi pembelajaran berbasis masalah lebih tepat daripada model pembelajaran ekspositori dalam meningkatkan hasil belajar ekonomi siswa, dan siswa yang memiliki tipe kepribadian ekstrovert akan memperoleh hasil yang lebih baik dari pada siswa yang memiliki tipe kepribadian introvert. Kata Kunci: strategi pembelajaran, tipe kepribadian, hasil belajar ekonomi. Abstract: This study aims: (1) To find out the results of students' economic learning taught by problem-based learning strategy is higher than students who are taught by expository learning strategy. (2) To know the economic learning result of students who have extrovert personality type and students who have introverted personality. (3) To know the interaction between learning strategy with personality type to Economic learning result. The research method used is quasi experiment with 2 x 2 factorial design. Statistical test used is descriptive statistics to present the data and continued with inferential statistic by using two way ANOVA with significant level α = 0,05 followed by Scheffe test. The results showed: (1) the students 'economic learning outcomes taught with problem-based learning strategy is higher than the students' economic learning outcomes taught with expository learning strategies; (2) the students 'economic learning outcomes that have extroverted personality is higher than the students' economic learning outcomes that have introverted personality types; (3) there is interaction between learning strategy with personality type in influencing student learning outcomes. This hypothesis suggests that problem-based learning strategies are more appropriate than expository learning models in improving students' economic learning outcomes, and students with extroverted personality types will achieve better outcomes than students with introverted personality types. Keywords: learning strategy, personality type, economic learning result

Levenshtein Augmentation Improves Performance of SMILES Based Deep-Learning Synthesis Prediction

2020 ◽  
Author(s):  
Dean Sumner ◽  
Jiazhen He ◽  
Amol Thakkar ◽  
Ola Engkvist ◽  
Esben Jannik Bjerrum
Keyword(s):  
Neural Networks ◽  
Pattern Recognition ◽  
Deep Learning ◽  
Recurrent Neural Networks ◽  
Data Augmentation ◽  
State Of The Art ◽  
Sequence Similarity ◽  
Learning Models ◽  
Underlying Network

<p>SMILES randomization, a form of data augmentation, has previously been shown to increase the performance of deep learning models compared to non-augmented baselines. Here, we propose a novel data augmentation method we call “Levenshtein augmentation” which considers local SMILES sub-sequence similarity between reactants and their respective products when creating training pairs. The performance of Levenshtein augmentation was tested using two state of the art models - transformer and sequence-to-sequence based recurrent neural networks with attention. Levenshtein augmentation demonstrated an increase performance over non-augmented, and conventionally SMILES randomization augmented data when used for training of baseline models. Furthermore, Levenshtein augmentation seemingly results in what we define as <i>attentional gain </i>– an enhancement in the pattern recognition capabilities of the underlying network to molecular motifs.</p>

Sign in / Sign up

Export Citation Format

Share Document