CS223 · Annotated SlidesChapter 6 — Orthogonal Projections, Gram–Schmidt and QR
Chapter 6 · Annotated slides

Orthogonal Projections, Gram–Schmidt and QR

Orthogonal projection onto a subspace ŷ = Σ (y·uᵢ)/(uᵢ·uᵢ) uᵢ, the orthogonal decomposition y = ŷ + z, the best approximation theorem, the Gram–Schmidt process and the QR factorization.

Slide text Study notes Highlight: key words Tips, tricks & traps Seen again elsewhere How it can come up

2
Exam weight Low
    Slide 3: Slide 3
    Exam weight Low
      Slide 4: Slide 4
      Exam weight High
      1. Gram–Schmidt: v₁ = x₁, v₂ = x₂ − (x₂·v₁)/(v₁·v₁) v₁, v₃ = x₃ − proj onto v₁ − proj onto v₂ … then normalize for an orthonormal basis.
      🎯 How this can come up · 1 question
      Written

      Apply Gram–Schmidt to x₁ = (3, 6, 0), x₂ = (1, 2, 2).

      Slide 5: Slide 5
      Exam weight Low

        6
        Exam weight Low
          Slide 7: Slide 7
          Exam weight Low
          1. QR: A = QR, Q has orthonormal columns (Gram–Schmidt on A’s columns), R = QᵀA is upper triangular with positive diagonal.
          Slide 8: Slide 8
          Exam weight Low