An improved learning rule for a neuron controller in a quick locating system

Author(s):  
Song Yibin ◽  
Wang Peijin
2020 ◽  
Author(s):  
Yuhan Helena Liu ◽  
Stephen Smith ◽  
Stefan Mihalas ◽  
Eric Shea-Brown ◽  
Uygar Sümbül

AbstractAnimals learn and form memories by jointly adjusting the efficacy of their synapses. How they efficiently solve the underlying temporal credit assignment problem remains elusive. Here, we re-analyze the mathematical basis of gradient descent learning in recurrent spiking neural networks (RSNNs) in light of the recent single-cell transcriptomic evidence for cell-type-specific local neuropeptide signaling in the cortex. Our normative theory posits an important role for the notion of neuronal cell types and local diffusive communication by enabling biologically plausible and efficient weight update. While obeying fundamental biological constraints, including separating excitatory vs inhibitory cell types and observing connection sparsity, we trained RSNNs for temporal credit assignment tasks spanning seconds and observed that the inclusion of local modulatory signaling improved learning efficiency. Our learning rule puts forth a novel form of interaction between modulatory signals and synaptic transmission. Moreover, it suggests a computationally efficient on-chip learning method for bio-inspired artificial intelligence.


2019 ◽  
Vol 3 (1) ◽  
pp. 197
Author(s):  
Rosita L. Tobing

The problem of classroom action research is the low learning outcomes of VC grade 164 students in Pekanbaru. This study aims to improve social studies learning outcomes of VC grade 164 students in Pekanbaru by applying the cooperative method of numbered heads together (NHT). The results of the research and class actions of the Social Studies Course conducted at the VC class SDN 164 Pekanbaru students concluded; Learning outcomes in the first cycle have increased compared to conventional learning. Pre-cycle learning outcomes are an average of 50.25 or sufficient categories; in cycle I, learning outcomes reached an average of 71.75 or in the Good category; in cycle II it increased again by 80.25 or in the Good category; Prasiklus classical completeness is 10 students (25.00%.); the first cycle is 27 students (67.50%); and in the second cycle were 38 students (95.00%). Students who have not been completed are remedial. Observers observed that VC grade 164 students at Pekanbaru Pekanbaru seemed to understand the Numbered Heads Together (NHT) Cooperative Method. They learn and understand shared material in heterogeneous groups of 4-5 students. Based on the results of improved learning studies, the application of the cooperative method of numbered heads together (NHT) succeeded in correcting the problem of the low social studies learning outcomes in VC Class SDN 164 Pekanbaru 2017/2018 Academic Year.


2018 ◽  
Vol 2 (3) ◽  
pp. 444
Author(s):  
Fuji Nengsih

IPS learning is a science of socio-cultural phenomena, and economics. IPS education in primary schools aims todevelop student potential. This study is a classroom action research that aims to improve the learning processwith the ultimate impact of improved learning outcomes. Data obtained on teacher activity cycle II percentage62.5% and 71% at the second meeting. Cycle II the percentage of teacher activity 83% and 92% at the secondmeeting whereas in student activity on cycle I with percentage 50% and second meeting 62,5% increase in cycleII become 75% and 88% at second meeting cycle II. The activity of teachers and students influences the IPSlearning result data with average views on the initial data 68.3, increased to 79.8 and in the daily test II with anaverage of 89.5. The conclusions in this study are make-match strategies effective in improving IPS learningoutcomes.


2018 ◽  
Vol 2 (3) ◽  
pp. 395
Author(s):  
Burhanuddin Burhanuddin

This research is motivated by the learning result of Social Sciences of Grade VI SD Negeri 022 Jaya Mukti KotaDumai which is still very low. This study aims to improve the learning outcomes of Social Sciences students.From the data analysis there is an increase of both teacher activity, student activity, and student learning result,that is teacher activity at meeting 1 cycle I percentage is 65% (enough) and at meeting 2 increase to 80%(good). In the second cycle of meeting 3 it increases again to 90% (very good) and at meeting 4 increases to95% (very good). Judging from the student activity also increased from the 1st meeting of cycle I was 60%(enough) and at meeting 2 increased to 70% (good). In the second cycle of meeting 3 it increased to 85% (verygood) and at the 4th meeting to 95% (very good). Judging from student learning outcomes, the average basicscore 63 increased to 75 in the first cycle of increase 12 points later in cycle II increased to 95 in cycle II largeincrease of 20 points. From the data analysis there is an increase both from teacher activity, student activity,and student learning outcomes. It can be concluded that the Improved Learning Model concept map can improvethe learning outcomes of IPS students of class VI SD Negeri 022 Jaya Mukti Kota Dumai.


2020 ◽  
Vol 16 (2) ◽  
pp. 280-289
Author(s):  
Ghalib H. Alshammri ◽  
Walid K. M. Ahmed ◽  
Victor B. Lawrence

Background: The architecture and sequential learning rule-based underlying ARFIS (adaptive-receiver-based fuzzy inference system) are proposed to estimate and predict the adaptive threshold-based detection scheme for diffusion-based molecular communication (DMC). Method: The proposed system forwards an estimate of the received bits based on the current molecular cumulative concentration, which is derived using sequential training-based principle with weight and bias and an input-output mapping based on both human knowledge in the form of fuzzy IFTHEN rules. The ARFIS architecture is employed to model nonlinear molecular communication to predict the received bits over time series. Result: This procedure is suitable for binary On-OFF-Keying (Book signaling), where the receiver bio-nanomachine (Rx Bio-NM) adapts the 1/0-bit detection threshold based on all previous received molecular cumulative concentrations to alleviate the inter-symbol interference (ISI) problem and reception noise. Conclusion: Theoretical and simulation results show the improvement in diffusion-based molecular throughput and the optimal number of molecules in transmission. Furthermore, the performance evaluation in various noisy channel sources shows promising improvement in the un-coded bit error rate (BER) compared with other threshold-based detection schemes in the literature.


Sign in / Sign up

Export Citation Format

Share Document