Đại học Kinh tế Quốc dân – National Economics UniversityMICRO4ASIA
🏆 Arena
Sign inSign up
MICRO4ASIAThe six micro-credentials

Courses are delivered on the Limio learning platform. Limio

← Mathematical Foundations for Data Science

You are viewing a public preview

Sign in to save your progress, do assignments, join discussions and get feedback.

Sign in
Part II — Calculus and Optimization

Topic 6 · Functions, Exponential and Logarithmic Functions

Print / Save as PDF
Lecture video — Topic 6: Functions, Exponential and Logarithmic Functions

Functions and model parameters, affine and linear functions, exponentials, logarithms and log transformations.

Flashcards · Functions, exponentials and logarithms

9 words
WordMeaningExample
  • Function, domain, range

    A function assigns exactly one output to each input in its domain. The range is the set of outputs actually taken.

    f(x) = x² on ℝ has range [0, ∞).

  • Natural real domain

    The real inputs for which a formula is defined.

    ln(x − 2) needs x − 2 > 0, so the domain is (2, ∞).

  • Affine function

    f(x) = b + wx: intercept b and constant slope w, so f(x + h) − f(x) = wh at every x.

    In ML, b + wx is called a linear model.

  • Exponential and its range

    eˣ has domain ℝ and range (0, ∞): always positive, never 0.

    3 + e^(−x) has range (3, ∞).

  • Natural logarithm

    ln is the inverse of eˣ: ln y = x ⇔ eˣ = y for y > 0. Also ln(eˣ) = x, ln 1 = 0, ln e = 1.

  • Three log rules

    ln(uv) = ln u + ln v; ln(u/v) = ln u − ln v; ln(uᶜ) = c ln u. There is no rule for ln(u + v).

    ln(2e³) = ln 2 + 3.

  • Exponential growth on the log scale

    A(t) = A₀e^(kt) gives ln A(t) = ln A₀ + kt, a straight line with slope k. Doubling time = ln 2 / k.

    A rise of 10% per year has k = ln 1.1 ≈ 0.0953.

  • Sigmoid

    σ(z) = 1/(1 + e^(−z)) maps any score to a probability strictly between 0 and 1; σ(0) = 1/2 and σ is increasing.

  • Odds and log-odds

    odds = p/(1 − p). The log-odds ln(p/(1 − p)) is the inverse of σ. One unit of score multiplies the odds by e.

    p = 3/4 has odds 3 and score ln 3.

Practice in Python

Work through Section 1 of Lab 2 (Calculus and Optimization, MFDS_Lab2_Calculus_Optimization.ipynb), attached to the Part II materials lesson at the start of this part.