Uniform Distribution
What you'll learn: How to work with uniform distributions where every outcome has an equal probability of occurring.
What Is a Uniform Distribution?
A uniform distribution is the simplest probability distribution where all outcomes are equally likely. Imagine rolling a fair die—each face (1 through 6) has exactly the same chance of appearing. That's a uniform distribution in action!
Discrete Uniform Distribution
In the discrete case, you have a finite set of outcomes, each with equal probability.
For a discrete uniform distribution over n values:
- Probability Mass Function:
P(X = x) = 1/nfor each possible value - All outcomes have identical probability
- Example: Rolling a fair die has probability
1/6for each face
Mean: The average of all possible values
Variance: Measures spread; for integers 1 to n, variance = (n² - 1)/12
Continuous Uniform Distribution
In the continuous case, any value within an interval [a, b] is equally likely.
- Probability Density Function:
f(x) = 1/(b-a)fora ≤ x ≤ b, and 0 elsewhere - Think of spinning a perfect spinner that can land anywhere between 0° and 360°
- Example: Randomly selecting a number between 0 and 1
Mean: (a + b)/2 (the midpoint)
Variance: (b - a)²/12
Why It Matters in ML
Uniform distributions are crucial for:
- Random initialization of neural network weights
- Random sampling from datasets
- Modeling scenarios with no prior knowledge (maximum uncertainty within bounds)
Key Takeaway: Uniform distributions assign equal probability to all outcomes (discrete) or all intervals (continuous), making them the "fair" baseline distribution when all possibilities should be treated equally.