CS220P Data Analytics, KAUST

Week 1 labs

Four interactive labs. Each one walks through the lecture's ideas step by step with live parameters, and ends with a tournament where every student submits exactly one official run. Each key step also has a collapsible graduate-level section.

Thursday, lab 1 · Intelligent agents

Keep the servers cool, cheaply

A data-centre cooling controller: reflex, model-based, goal-based, utility-based and learning agents, then the cooling league.

Open the lab

Thursday, lab 2 · Solving problems by searching

The campus delivery robot

BFS, DFS, iterative deepening, uniform-cost, greedy and A* on a campus map, then the dispatch league.

Open the lab

Sunday, lab 1 · Data mining and data

The student survey

Attribute types, measurement error, outliers, missing values, duplicates, sampling, transformations, PCA, then the cleaning challenge.

Open the lab

Sunday, lab 2 · Measures of similarity

Find your match

Minkowski distances, standardisation, SMC vs Jaccard, cosine for documents, correlation, Mahalanobis, then the matchmaker.

Open the lab

How the tournaments work

  1. The TA announces a room code per table or team (for example A1, A2, B1).
  2. Students open the lab's tournament step, enter their name and room, and use Practice run as often as they like. Practice uses practice data; the official run uses different hidden data, so the winning strategy is the robust one.
  3. Official run works once per device and produces a submission code with a checksum.
  4. Students send the code to the TA (chat, form, or read it out).

Everything runs in the browser. No installation, no account, works on phones.