Forums
New posts
Search forums
What's new
New posts
New media
New media comments
Latest activity
Media
New media
New comments
Search media
Members
Registered members
Current visitors
Log in
Register
What's new
Search
Search
Search titles only
By:
New posts
Search forums
Menu
Log in
Register
Install the app
Install
FJR Rides and Gatherings
Long Distance Riding/Iron Butt Rally
Traveling Salesman algorithm with simulated Annealing
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser
.
Reply to thread
Help Support Yamaha FJR Motorcycle Forum:
This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Message
<blockquote data-quote="El Toro" data-source="post: 1277620" data-attributes="member: 14779"><p>Interesting link.</p><p></p><p>My take on it is that the concept of "naive hill climbing" is at best ... naive. Using naive hill climbing as a basis for comparison puts the simulated annealing in the best possible light.</p><p></p><p>When people cared about searching for optima by clever strategies (back in the days before exhaustive search became feasible for so many problems), heuristics were applied to keep out of cul de sacs, or to avoid local minima. One method was to use a shotgun approach to find starting points, or to choose starting points that were clearly separated. No guarantees, of course. But similated annealing doesn't offer any guarantees either.</p><p></p><p>FWIW, there are lots of ways to solve this problem.</p><p></p><p>Note that their initial route is a foolish route that only an incredibly stupid person would have chosen "at random."</p><p></p><p>Note that their final route is one that a good travelling salesman might have been quite close to by just using the strategy of finding the next nearest stopping point in the general direction of crossing the country from the current point.</p><p></p><p>Simulated annealing has been around for a long time now .... This algorithm might be useful for trip planning if the trip is strongly time limited, with regular changes in target destinations depending on events en route. At the worst, it would provide comfort and relief to the rider. At the best, it could offer a competitive advantage for the user.</p><p></p><p>Thanks for sharing the link.</p></blockquote><p></p>
[QUOTE="El Toro, post: 1277620, member: 14779"] Interesting link. My take on it is that the concept of "naive hill climbing" is at best ... naive. Using naive hill climbing as a basis for comparison puts the simulated annealing in the best possible light. When people cared about searching for optima by clever strategies (back in the days before exhaustive search became feasible for so many problems), heuristics were applied to keep out of cul de sacs, or to avoid local minima. One method was to use a shotgun approach to find starting points, or to choose starting points that were clearly separated. No guarantees, of course. But similated annealing doesn't offer any guarantees either. FWIW, there are lots of ways to solve this problem. Note that their initial route is a foolish route that only an incredibly stupid person would have chosen "at random." Note that their final route is one that a good travelling salesman might have been quite close to by just using the strategy of finding the next nearest stopping point in the general direction of crossing the country from the current point. Simulated annealing has been around for a long time now .... This algorithm might be useful for trip planning if the trip is strongly time limited, with regular changes in target destinations depending on events en route. At the worst, it would provide comfort and relief to the rider. At the best, it could offer a competitive advantage for the user. Thanks for sharing the link. [/QUOTE]
Insert quotes…
Verification
Post reply
FJR Rides and Gatherings
Long Distance Riding/Iron Butt Rally
Traveling Salesman algorithm with simulated Annealing
Top