Programming (Leetcode)
Return sum of all even numbers in a list
Fill in the blank:
def sum_of_evens(numbers):
return __
sum(num for num in numbers if num % 2 == 0)
What term describes AI systems that improve on performance through learning from data?
Machine Learning
Name any web browser created in the 1990s that had contemporary mass-market success.
Double points for answering with the first mass-market browser.
Other answers: Lynx, Netscape Navigator, Opera, Internet Explorer, Netscape Communicator
Input, hidden, output are examples of something that makes up neural networks. What is it called?
Layers,
Gates is also a close answer (in LSTM) but not entirely correct due to forget gate instead of hidden layer.
What is the definition of a Set in mathematics?
A collection of distinct objects
Alt: A collection of different things
Reverse a string
Fill in the blank
def reverse_string(s):
return __
s[::-1]
This test determines if a machine can mimic human conversation well enough to fool a human judge.
The Turing Test
Which three computer models comprised the first generation of mass-market personal computers?
They were referred to as the Holy Trinity, and were from three different companies. Normal points for getting 1, double points for getting 2
Apple II, the Commodore PET, and the TRS-80
This iterative method updates model parameters by moving in the direction opposite to the gradient of the loss.
Gradient Descent.
What is it called when a compound statement is always true no matter what the individual parts consist of?
Write a Python function that takes a list of integers and returns a new list containing only the unique elements (no duplicates).
Fill in the blank:
def unique_elements(numbers):
return __
list(set(numbers))
When an AI system makes hard-to-explain decisions, what quality is it lacking?
Interpretability or transparency.
What does HTTP stand for?
Hypertext Transfer Protocol
What is the term for the process that adjusts the weights of a neural network based on the error of the output compared to the expected result?
Backpropagation
What is the probability of rolling at least one 6 (six) with 5 (five) dice rolls of a fair die? Calculators allowed.
1 - (5/6)5
= 0.59812242799
Write a Python function to implement the quicksort algorithm, which sorts an array of integers in ascending order.
Fill in the blank:
def quicksort(arr):
if len(arr) <= 1:
return arr
pivot = arr[len(arr) // 2]
left = [x for x in arr if x < pivot]
middle = [x for x in arr if x == pivot]
right = [x for x in arr if x > pivot]
return __
quicksort(left) + middle + quicksort(right)
This technique allows a model trained on cats to be reused for hamster recognition, for example.
Transfer Learning
What was the name of the first artificial neural network, developed in 1958 by Frank Rosenblatt, that was designed to recognize patterns?
Perceptron
This ensemble method combines multiple decision trees using bootstrapped samples and averaging.
Double Jeopardy! Double the points if you answer correct, but also double the incorrect penalty.
Random Forest
How many edges are in a bipartite graph K3,4 ?
12 edges.
Write a Python function that implements a binary search algorithm on a sorted array, returning the index of the target value if found, or -1 if not found.
Fill in the blank:
def binary_search(arr, target):
left, right = 0, len(arr) - 1
while left <= right:
__
if arr[mid] == target:
return mid
elif arr[mid] < target:
left = mid + 1
else:
right = mid - 1
return -1
mid = left + (right - left) // 2
What is the paradox that hypothesizes that human tasks requiring intuition (such as walking) are harder for AI than abstracted tasks such as calculus?
Moravec's Paradox
Explanation: Sensorimotor skills (e.g., navigating a room) demand immense computational power, while abstract reasoning (e.g., chess) is relatively easier to replicate. It underscores why robotics and embodied AI lag behind symbolic AI systems.
Half points for one, full points for both
Apple Lisa (1983), Apple Macintosh (1984)
What is the tradeoff between a model's ability to fit training data versus generalizing to new data?
The bias-variance tradeoff
In a dataset with 100 points, a linear regression model has 5 features. How many degrees of freedom for the residuals in this model?
95