Search vs planning in artificial intelligence

I m studying artificial intelligence in Russell & Norvig book. We did a search and planning part that for me is the same (at least on the rappresentation). I wish know that there is a difference between this two technique, but I didn’t hit.

Looking online i found this resource: Planning and Search

where say “The main difference between search and planning is the representation of states”

Possible that the difference between this two technique is just the representation?

Scaled Agile Framework, PI Planning & Iteration Plans [on hold]

A read of the SAFe PI Planning documentation starts strong with clarifying that PI Planning produces milestones, features and dependencies and places these on the program board. In contrast, “Iteration Plans” or user stories are not listed as an output of PI Planning.

However, towards the end of that same link it makes a passing comment of when PI Planning completes the teams “take PI objectives, iteration plans, and risks back to their regular work area.” Furthermore, the SAFe documentation for Iteration Planning makes the following comment:

In SAFe, iteration planning is a refinement of the level of detail and an adjustment of the initial iteration plans created during Agile Release Train (ART) PI planning.

This SAFe commentary is much too thin to draw conclusions about what was the activity and what was the goal of creating iteration plans during PI Planning.

My interpretation is that any team might create loose iteration plans during PI Planning, but only with the goal of estimating when they will complete a given feature(s). In other words, such loose iteration planning (if it happened at all) is a team-internal scratch-pad that would not be evaluated or published. Accordingly the program board would not be decorated with user stories that result from such loose iteration planning (and in general the program board simply does not have user stories on it).

In my interpretation, no agile coach would tell my team “your scratch-pad iteration plans are insufficient, you must create more.”

In contrast, I recently encountered an organization whose interpretation of SAFe is as follows:

  • SAFe PI Planning requires or expects the activity of Iteration Planning that intentionally covers each and every sprint of the upcoming Program Increment (which is 3 months in their case).
  • The user stories that are created from this comprehensive whole-PI “Iteration Planning,” are placed onto the program board (as a backbone of creating dependencies with features, etcetera).
  • At the end of PI Planning, if the entire Program Increment does not have end-to-end Iteration Plans established for a given team, then that misaligned team must ASAP finish a full suite of Iteration Plans for the entire PI.

In this latter view, subsequent sprint planning for each iteration involves merely making adjustments to the sprints if needed – because otherwise all sprints of the PI have already been planned.

Please note that the above bulleted premises are not seen as customizations of SAFe, but rather they believe it represents the authentic interpretation of SAFe.

I’d like to know if SAFe provides any additional guidance that would clarify which interpretation is “correct.” To me, which interpretation is chosen is a choice between genuinely following an Agile methodology versus following a practice that is half-way waterfall. However, my question is not “what do you prefer,” as what I’m really after is a more official or authentic interpretation of SAFe on this matter.


Is there a reason to use specifically fibonacci sequence in planning poker?

I’ve noted that fibonacci sequence is quite popular in planning poker, but is it a reason for that particular sequence? Wouldn’t for example powers of 2 work equally well?

Both sequences are more or less exponential while fibonacci uses a factor of the golden ratio (approximately 1.6) so fibonacci has somewhat higher resolution and would allow to express more accurate estimates.

Is there for example any evidence that people tend to be able to estimate accurate enough to motivate the higher resolution? And if there is wouldn’t a even finer scale be motivated?

cheap multi agent path planning algorithms

I am trying to make a tower defence game where some critters are trying to make their ways through a 2D grid-like obstacle course. The idea is these critters cannot walk into the obstacles or into each other during their path, and they would like to get to the other side as soon as possible.

I just need a cheap, not-optimal solution that can path all the agents from one side of the map to another, moving in the grid (up-down-left-right) directions while not walking into each other nor into the obstacles.

I am not wanting an optimal solution, just one that looks reasonably greedily optimal would be nice (i.e. not like not moving the 2nd agent without completing the path on the first agent, etc).

I’m thinking about something with potential fields with a little bit of A* is as much as I am willing to compute. If there’s anything simple to implement I’d be happy to know ! !

I am planning to dual boot my computer with Windows 10 and Ubuntu on Legacy mode

which should I use for legacy Rufus or Universal USB installer? Or does it even matter? I had a dual boot system of ubuntu and 8.1 but I fresh installed windows 10(used rufus to create bootable usb selecting MBR not GPT) and the system boots straight to windows (no GRUB menu), so i think my ubuntu was in uefi or something so I am planning to fresh install Ubuntu as well..

Public transport planning in Spain

Is there a comprehensive public transport planner for Spain?

From what I found so far it looks like some companies have their own planner, but obviously that wouldn’t work when traveling with more than one company.

There is for instance, but it seems to be usable for direct connections only. An example: I can find both San Sebastian → Bilbao and Bilbao → Haro but not San Sebastian → Haro (either direct or via Bilbao).

I’m specifically interested in the Basque and Rioja regions. So, if no planner exists, it would also help to know which companies are active there.

Self-organization during Sprint Planning

During Sprint Planning there are a lot of decisions to be made:

  1. How many PBIs should the team commit to?
  2. Which tasks should constitute each PBI?
  3. How much time should each task take?

With no team leader it is not always easy to make all these decisions, since there will be some decision points (sometimes more than a few) in which the team members will disagree.

I thought about a few options to deal with such disagreements when a consensus cannot be reached in a short period of time (we don’t want endless arguments).

  1. Let ScrumMaster decide
  2. Let the person who will most likely work on the story decide.
  3. Majority vote
  4. Let the person who actually writes the tasks in Excel decide (in such case have a different person write the tasks for each story).

I have two questions:

  1. How to measure consensus? Should the Scrum Master ask each person whether they agree with every decision proposed?

  2. When no consensus can be reached shortly, which option from the list that I proposed do you think we should choose, if any?

Thank you!

In scrum, if tasks are estimated in hours, how to avoid assigning the task in sprint planning?

I’ve been reading up a lot on the subject, and from what I can gather the following statements ‘should’ be true in scrum:

1) Estimate user stories in story points at the product backlog level. This allows abstraction of the estimates, and the team can estimate together without dispute – e.g. one developer may complete the story much quicker than another, but for both, the estimate is 3 story points.

2) You ideally shouldn’t assign tasks during sprint planning. While it helps with indivdual accountability, it reduces team accountability

3) In sprint planning, flesh out the tasks involved for each story, and estimate these in hours.

My question is this, how can you estimate the tasks in hours, yet still not assign the task to a specific developer (points 2 and 3)? Surely the whole point of using story points is to avoid the expert developer saying it’ll take 1 hour, and the junior dev saying it will take 8 hours.

If you are now estimating tasks within sprint planning in hours, how do you reach an estimate which is good for all team members of varying levels of expertise, without assigning the task to the specific dev who’s estimated it?