WIR-020
Recurring commitments
Physical & Mental Health
| Task | Frequency | Status |
|---|---|---|
| Exercise 30+ minutes. | x5/week | ❌ |
| Wake before 08:00. | x4/week | ✅ |
| Meditate 3+ minutes. | x5/week | ❌ |
- Exercise 30+ minutes
- Ran 3/5 days this week. Two easy runs (one long), and one relatively short tempo.
- Wake before 08:00
- Nailed this goal; 6/7 days.
- Meditate 3+ minutes
- Completely neglected this task. 1/7 days.
Relationships
| Task | Frequency | Status |
|---|---|---|
| Contact 2+ family members/day. | daily | ❌ |
| Visit 1+ new place in town/week. | x1/week | ✅ |
- Contact 2+ family members/day
- Close, but we’re aiming for 7/7 days here. 5/7 days.
Research
| Task | Frequency | Status |
|---|---|---|
| Submit a short conceptual post to the website; cover a topic in math/AI. | x1/week | ❌ |
| Complete 1+ deep work session. | x6/week | ❌ |
- Submit a short conceptual post to the website; cover a topic in math/AI
- Currently cooking up a lovely blog post on back-propagation for you folks. Still deep in the pipeline unfortunately.
- Complete 1+ deep work session
- It looks like we marked 4/7 days complete on our task planner. We probably did better than this and simply forgot to record it. Will not be awarding any partial credit here.
Previous week’s action items
- 1. Add a little bit to WIR at the end of each workday.
- 2. [MON] Update the personal constitution; update this week’s action items thereafter.
- 3. Wake up at 8:00AM (4/7) days.
Work completed
| Date | DOTW | Work Completed |
|---|---|---|
| 03-01-26 | Sunday | [LOG] WIR + correspondence [RSR] PIML-QPE |
| 03-03-26 | Monday | [RSR] PIML-QPE – data analysis; dataset updates |
| 03-04-26 | Tuesday | [RSR] PIML-QPE – data analysis; slides/comms |
| 03-05-26 | Wednesday | [RSR] PIML-QPE – debugging, dataset + task-formulation |
| 03-06-26 | Thursday | Gone storm chasin’ ⛈️ |
| 03-07-26 | Friday | [PhD] Interviews/meetings |
| 03-08-26 | Saturday | [RSR] PIML-QPE – error/bias correction baselines/experiments |
Reflections
The week-in-review series serves two purposes: to raise the ceiling of our good weeks, and to raise the floor of our bad weeks. These past two weeks have certainly fallen into that latter category. We’ve done a poor job of sticking to the recurring commitments set out at the beginning of the year, and we’re still lacking concrete research goals at the weekly and monthly scales to adhere to.
This combination of structural weaknesses creates the following behavioral loop.
- You feel ambitious –> begin working on an ambitious task –> encounter resistance –> give up and sulk
Ideally, we want to modify this chain of events like so.
- You feel ambitious –> temper your ambitions and make reasonable plans –> encounter resistance –> persevere and succeed
The good news is that people are highly adaptive, and admitting one’s flaws is always the first step on the path to better habits. Next week we’re going to make a couple of adjustments, starting in the goal setting department. For all targets we failed to hit, we’ll go ahead and update task frequency and difficulty. Second, we’re going to officially commit to daily check-ins. For tasks like “deep work” and “contact friends/family” (i.e., daily, recurring tasks), it will be much easier to build up the habit of adding to the WIR each day. After all, procrastination simply isn’t an option for recurring commitments, and therefore we’ll need to implement other systems.
Lastly, we’ll want to plan on sharing regular research reports. Our current weekly/bi-weekly meeting structure provides too sparse of feedback on projects. By getting in the habit of publicly documenting progress on projects, we are more likely to improve organization and accountability with respect to academic responsibilities. For my rabid fanbase of readers, this will be the perfect opportunity to scoop some truly high-impact work.
Action items
- 1. Complete daily WIR check-ins
- At end of each day, post progress on all recurring commitments to blog
- 2. Update personal constitution
- Modify all failed goals for 03/08
- Segment weekly AI/math blog post goal into smaller chunks
- Add: 2-3x research reports/week
- Modify all failed goals for 03/08
- 3. Complete 2+ research reports + add to blog
- 4. Complete backprop post + add to blog
Until next week 👋
Enjoy Reading This Article?
Here are some more articles you might like to read next: