Post your solution here

-- Practice Final

Problem-4 In our categories and objects approach to knowledge representation show how the BunchOf predicate can be defined in terms of PartOf predicate. Name: Harita Shroff,Archana Yadawa,Changtong Zhou, Bundit Hongmanee, Jonathan Neel ((resource:FullSizeRender (1).jpg|Resource Description for FullSizeRender (1).jpg))

-- Practice Final

 Group Members: Phyllis Lau, Justin Ng, Wei Su
 Problem 3:

Group Members: Phyllis Lau, Justin Ng, Wei Su Problem 3: ((resource:IMG_5831.JPG|Resource Description for IMG_5831.JPG))

-- Practice Final

Problem 10. Team: Daniel Vu, Derek Ortega, Alec Leong K-nearest neighbors algorithm: Suppose when given a query x⃗ q, we lookup the k nearest neighbors to x⃗ q with respect to some kind of distance function.Denote these neighbors by NN(k,x⃗ q).To classify x⃗ q, we take the majority vote of these k neighbors. A nonparametric model is one that cannot be characterized by a bounded set of parameters. This is nonparametric as that we need to keep all of the training data in order to run this algorithm -- we don't learn some fixed parameters and then forget the training data.

-- Practice Final

2. Consider the task of taking out the garbage, this might involve sorting refuse into garbage or recycle piles, putting piles into garbage or recycle bins, and taking each bin out to the curb. Model this problem as a PDDL planning problem. Give an example solution plan. Students: Todor, Naomi, Jerry, Noor ((resource:Screenshot_2017-12-11_15-54-22.png|Resource Description for Screenshot_2017-12-11_15-54-22.png))

-- Practice Final

8. Team: Rui Li, Michael Torres ((resource:8.jpg|Resource Description for 8.jpg))

-- Practice Final

'''''Oindril Dutta, Bao Pham, Sanjana Shetty''''' ---- '''#5. Consider the following rules:''' ---- * Human(x) ^ tall(x) : giant(x) | giant(x) * Human(x) ^ !tall(x) : short(x) | short(x) * Human(x) ^ wide(x) : fat(x) | fat(x) * Human(x) ^ !wide(x) : thin(x) | thin(x) ---- '''With the following facts:''' ---- * Human(Bob) * tall(Bob) * !tall(Bob) * wide(Bob) * !wide(Bob) ---- '''Gives the following four extensions:''' ---- # {Human(Bob), tall(Bob), giant(Bob)} # {Human(Bob), !tall(Bob), short(Bob)} # {Human(Bob), wide(Bob), fat(Bob)} # {Human(Bob), !wide(Bob), thin(Bob)}

-- Practice Final

Team: Wenxiang Hu, Yecheng Liang. Problem 7. ((resource:IMG_2948.JPG|Resource Description for IMG_2948.JPG))

-- Practice Final

Alexander Duong Stephen Reyes #9. Give an example of a function not computable by a single perceptron gate. Show that it can be computed by a multi-layer perceptron network. The XOR function is an example of a function not computable by a single gate because outputs cannot be separated linearly. To show that the XOR function can be computed with a multi-layer perceptron network, we can show the scatteplot of an XOR function. The graph is not linearly seperatable and needs at least 2 lines to separate the values. By adding more layers, we can create another linear division that further separates the categories.

-- Practice Final

Group #1: Rodion Yaryy, Sridivya Kondapalli, Ross Wilkinson, Joy Yan. Question #1 ((resource:IMG_9933.JPG|Resource Description for IMG_9933.JPG))