Section 60.5: Lab infrastructure and compute budgeting for instructors

"My lab budget has three settings: local, cloud, and please use the small model today."

A Compute Plan With Receipts
Technical illustration for Section 60.5: Lab infrastructure and compute budgeting for instructors.
Figure 60.5A: Lab infrastructure budget tiers for instructors: a zero-hardware tier using free Colab GPUs and MuJoCo, a mid-tier using a small workstation cluster for IsaacGym, and a full-hardware tier with two real robot arms for capstone demos.
Big Picture

Lab infrastructure and compute budgeting for instructors gives Teaching with This Book a concrete systems role: design labs around predictable compute, reproducible environments, and graceful fallback paths. The section keeps asking what the agent observes, what it remembers or updates, which action changes, and what evidence would convince a skeptical reader.

This section develops the technical contract for lab infrastructure and compute budgeting for instructors into a usable mental model. First we define the object of study, then we connect it to the agent loop, then we test it with a compact implementation.

The key question in Lab infrastructure and compute budgeting for instructors is practical: what must the agent know, what can it observe, what action is available, and what evidence shows that the action worked under the stated conditions?

Action Is The Test

Lab infrastructure and compute budgets should be judged by the action it improves. A section claim is strong when it names the decision, the measurement, and the failure mode before a larger model or simulator is introduced.

Theory

For Lab infrastructure and compute budgeting for instructors, the practical design rule is to make the interface inspectable before optimization begins: inputs, outputs, units, latency, bounds, and failure labels should all be visible in the saved artifact.

Mechanism

The mechanism in Lab infrastructure and compute budgeting for instructors is the contract between representation and action. Name what enters the module, what leaves it, which assumptions make that transformation valid, and which log would reveal a bad handoff.

Worked Example

For Lab infrastructure and compute budgeting for instructors, keep one concrete rollout in view. A sensor reading becomes an estimate, the estimate constrains an action, the action changes the world, and the next observation confirms or contradicts the assumption. The section's idea is useful only if it improves that loop.

Library Shortcut

For Lab infrastructure and compute budgeting for instructors, the small contract exists to expose the teaching artifact before tooling takes over. Use notebooks, simulators, shared logs, rubrics, and capstone studios only when they preserve the same observation, action, metric, and failure fields.

Practical Recipe

  1. Write the observation, action, and success metric before choosing a model.
  2. Build a baseline that is simple enough to debug by inspection.
  3. Add the library implementation only after the baseline behavior is understood.
  4. Record failures as structured cases: perception error, state error, planning error, control error, or evaluation error.
  5. Run at least one perturbation test before trusting the result.
Common Failure Mode

The common mistake in Lab infrastructure and compute budgeting for instructors is to trust a component score before checking the closed-loop interface. The failure usually appears where state, timing, authority, or evaluation context crosses a module boundary.

Practical Example

A team using Lab infrastructure and compute budgeting for instructors starts by writing the task panel, not by picking the largest model. They keep a baseline run, a maintained-tool run, and a perturbation run in the same result folder. The comparison is accepted only when the action trace, metric, and failure labels come from one script.

Memory Hook

Treat lab infrastructure and compute budgeting for instructors like a control-room label. If the label does not tell a future debugger what moved, what sensed, or what failed, it is decoration rather than engineering knowledge.

Research Frontier

For Lab infrastructure and compute budgeting for instructors, the open research question is not whether a larger policy can produce a better demo. The sharper question is whether the method improves reliability across new scenes, new embodiments, delayed feedback, and rare failures under an evaluation protocol that another lab can reproduce.

Self Check

For Lab infrastructure and compute budgeting for instructors, can you name the observation, action, protected assumption, success metric, and one likely failure case? If any field is vague, rewrite the contract before adding model complexity.

Topic-Native Deepening

This section turns course logistics into a first-class design topic. Embodied AI courses fail unnecessarily when infrastructure assumptions remain implicit, such as hidden GPU requirements, fragile installs, or labs that have no fallback path for students with weak hardware.

The goal is not luxury compute, it is predictable compute. Instructors need a budgeting and fallback framework that keeps the course moving when a heavy job, a broken simulator, or a cloud quota limit appears in the middle of the semester.

Why This Section Matters

Lab infrastructure and compute budgeting for instructors becomes teachable once the student can state the operative variables, the decision boundary, and the evidence artifact. The section should therefore be read together with Chapter 11 on simulators and Chapter 59 on capstones, where the same loop is developed from adjacent angles.

Formal Object

For lab $i$, estimate total compute demand as $C_i = n_s(t_{cpu}+g_i t_{gpu}) + c_i t_{cloud}$, where $n_s$ is student count, $g_i$ is the fraction needing GPU, and $c_i$ is the fraction offloaded to cloud. The point is not exact accounting; it is seeing where bottlenecks can form before the assignment ships.

The budget model reveals which labs are fragile. A lab is risky when it requires most students to run long GPU jobs locally or when it has no cheap substitute that still teaches the mechanism.

Algorithm: Design a resilient lab infrastructure plan
  1. Classify each lab as CPU-safe, local-GPU, or cloud-first before the term begins.
  2. Provide one maintained environment per class, such as a pinned Colab or container image.
  3. Publish fallback routes, including smaller models, shorter horizons, or prerecorded logs for analysis.
  4. Track cloud budget, queue times, and peak usage weeks alongside assignment release dates.
  5. Run one dry-run audit on fresh machines before assigning the lab.
Infrastructure Planning Grid
DimensionWhat To SpecifyWhy It Matters
EnvironmentPinned notebooks, containers, simulator presetsPrevents environment drift.
Compute tierCPU, local GPU, or cloud offloadMakes hidden hardware assumptions visible.
Fallback pathSmaller model, shorter run, or analysis-only modeKeeps learning moving during outages.
Support artifactInstall log, runtime expectation, and dry-run screenshotReduces avoidable support load.

The expected output should let an instructor decide whether the lab is safe to assign. If no fallback path is visible, the lab is still operationally fragile.

Library Shortcut

After the from-scratch contract is clear, the practical route uses Colab, VS Code dev containers, Docker, MuJoCo Playground, cloud notebooks, GitHub Classroom, CI. The payoff is that standard interfaces, logging, batching, and replay support move from ad hoc glue code into maintained infrastructure, while the evidence schema stays the same.

Project Or Teaching Use

A practical rule is to make the most expensive labs optional extensions unless they teach a core mechanism that cannot be seen any other way. Students remember the concept, not the heroic setup time.

Research Frontier

The frontier teaching challenge is that embodied AI increasingly depends on heterogeneous infrastructure, from local simulators to cloud inference to robot teleoperation. Good course design must absorb that complexity without hiding it.

Expected Output Interpretation

For Lab infrastructure and compute budgeting for instructors, the artifact should show the course-design decision, the evidence students must produce, and the failure mode that would trigger a revised assignment or rubric.

Key Takeaway
Exercise 60.5.1

Design a method-matched experiment for Lab infrastructure and compute budgeting for instructors. Specify the environment, observation schema, action interface, metric, and one perturbation that targets the section's core assumption.

Section References

Biggs, J. Teaching for Quality Learning at University. Open University Press, 1999.

Use for constructive alignment between learning outcomes, activities, and assessment.

Anderson, L. W. and Krathwohl, D. R. A Taxonomy for Learning, Teaching, and Assessing. Longman, 2001.

Use for designing assessments that move from recall to analysis, creation, and evaluation.

What's Next?

Next, continue with the following teaching section, where the Lab infrastructure and compute budgeting for instructors contract becomes a concrete course-design decision.