FLIPPED CLASSROOM ACTIVITY CONSTRUCTOR – USING EXISTING CONTENT

Table of Contents SECTION SLIDE # 5 10 20 ABOUT YOU 3 OUT-OF-CLASS SEGMENT 5 IN-CLASS SEGMENT 10 EVALUATION 20 COMMUNITY BUILDING

About Me I am Vinod J. Kadam, Assistant Professor in Department of Information Technology, Dr.BATU, Lonere Maharashtra. My topic:-The McCulloch-Pitts model About Topic:- The McCulloch-Pitts model is an extremely simple artificial neuron. We see its equation and learning process. We will also learn to design simple Gates using this model

Vinod J.Kadam McCulloch Pitts Model Artificial Neural Network Computer Final Year UG students in IT Dr.Babasaheb Ambedkar Technological University

Out-of-class Activity Design-1 Learning Objective(s) of Out-of-Class Activity At the end of watching the videos student should be able to Explain simple McCulloch Pitts Model and it’s equation using graph Explain Linear threshold function using neat diagram. Design NOR gate and NAND gate using McCulloch Model. Key Concept(s) to be covered McCulloch Pitts Model and it’s equation Linear threshold function Graphical Representation of the model Simple neuron models: NOR gate NAND gate

Out-of-class Activity Design - 2 Main Video Source URL https://youtu.be/QSnSCNg7OMY License of Video CC-BY-SA (reuse allowed) Mapping Concept to Video Source CONCEPT VIDEO SEGMENT DURATION (in min) McCulloch Pitts Model and it’s equation V1 - 00:00 – 1:32 1.32 Linear threshold function V2 – 1:33 – 2.06 0.33 Graphical Representation of the model V3 – 2.07 – 3.13 1.06 Simple neuron models: NOR gate V4 - 3:14 – 4.00 0.46 Simple neuron models: NAND gate V5 – 4.01 – 4.48 0.47 TOTAL DURATION 5.29 min

Out-of-class Activity Design - 3 Aligning Assessment with Learning Objective Learning Objective Assessment Strategy Expected Duration (in min) Additional Instructions (if any) Understand simple McCulloch Pitts Model and it’s equation Q.1 .This model has ability to learn. True or False, Justify . 5 minutes Watch V1 and then answer Q1,Q2

Out-of-class Activity Design - 3 Aligning Assessment with Learning Objective Learning Objective Assessment Strategy Expected Duration (in min) Additional Instructions (if any) Explain Linear threshold function Q.1. What is nature of function F(x) in McCulloch Pits Models? Linear or Nonlinear. Explain. 5 minutes Watch V2 and then answer Q1

Out-of-class Activity Design - 3 Aligning Assessment with Learning Objective Learning Objective Assessment Strategy Expected Duration (in min) Additional Instructions (if any) Design NOR gate and NAND gate using McCulloch Model Q.1 If ‘b’ in the figure below is the bias, then what logic circuit does it represents? 5 minutes Submit answers to all questions 3 hours before coming to class. Total activity duration 15 minutes

In-class Activity Design -1 Learning Objective(s) of In-Class Activity At the end of the class, students will be able to, Design Memory Cell and basic Universal gates (AND,OR and NOT) by setting weights and thresholds Design Simple Model of Artificial Neural Network Key Concept(s) to be covered Use of McCulloch pitts Model for memory cell Learning process of Neural Network

In-class Activity Design -2 Active Learning activity(ies) that you plan to do Concept Clarification using. Think-Pair-Share Concept clarification using. Peer Instruction

In-class Activity Design -2 Peer Instruction Strategy – What Teacher Does Pose the three PI questions at the start of the class and provide summary of McCulloch Model and Explain learning process of Neural network. Q 1: Design AND and NOR gate using McCulloch Model.

In-class Activity Design -2 Peer Instruction Strategy – What Teacher Does Q 2: Design Memory Cell Using McCulloch Model.  

In-class Activity Design -2 Peer Instruction Strategy – What Teacher Does Q 3: If feedback is given to update weights, Can we improve the model?  

In-class Activity Design -2 Peer Instruction Strategy – What Student Does For each question they will first vote individually. Then they will discuss with peers and come to consensus. Listen to instructors explanation.

In-class Activity Design -2 TPS Strategy – What Instructor does The main “ideological” problems of the McCulloch-Pitts model were that the network must be completely specified before its using and there were no free parameters to suit different problems. Learning can only be implemented by modifying the connection pattern of the network and the thresholds of the units, which is necessarily more complex than just adjusting numerical parameters. We can modify this Mcculloch model for understanding simple learning process of Artificial Neural network.

In-class Activity Design -2 TPS Strategy – What Instructor does Think (~2 minutes) Instruction: consider following diagram and calculate output for Case 1:x1 is 0.8 and x2 is 0.3 Case 2:-x1 0.4 and x2=0.1 If target output for case 1 is 1 and that of case 2 is 0,then what should be weights?

In-class Activity Design -2 TPS Strategy – What Instructor does Pair (~5 minutes) Instruction: Now pair up and compare your answers. Agree on one final answer. While students are pairing and discussing, instructor goes to 2~3 sections to see what they are doing.

In-class Activity Design -2 TPS Strategy – What Instructor does Share (~8 minutes) Instructor asks a group to share their answer with class and see whether there are different answers. After sharing is done, instructor gives feedback on the correct solution. Instructor will explain weight updating concept Dwi = c(t – z) xi

In-class Activity Design -2 Justify why the above is an active learning strategy In both the above strategies, students are required to go beyond mere listening and execution of prescribed steps. They are required to think deeply about the content they were familiarized in out-of-class and do higher order thinking. There is also feedback provided (either through peer discussion or instructor summary)