h a l f b a k e r yIf you can read this you are not following too closely.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
AIs are hard to program because iteration will find paths
in
the data sets that, while being a solution, are not what is
wanted for the original problem. This is an opportunity
because these unique solutions are out side the box, new
fertile ground of possibility.
Take Cricket, young cricketers
watch the game , learn
the
game, get coached by past coaches, play the game for
years then maybe become coaches. This in itself is an
iterative process. Let's reduce the set and look at
batting.
I assume cricket techniques have been whittled out of
time
from different ball sport and martial art strokes but is
there a far, far out of the box batting technique human
cricketers have not substantiated? May be a AI can
crunch
some body and bat motions that our new cricketers can
use.
Model cricket for the standard bowling, field positions
but
let the human motion rig batter do want ever they want.
Weight for large hits, run shots. Run the cycles.
Who would ever thought a human could bat like that.
[link]
|
| |
The AI would wonder why so much time was
invested in such a dull sport... |
|
| |
[RS] Also an AI cannot do much useful work if the parameters are not clearly defined. You can set an AI to go through a data set if you give it a clear target "identify cats" and then grade its results on how many correct and incorrect solutions it gives. But it seems that the proposed set up has no such target behaviour, since playing cricket seems utterly pointless, even the scoring system is pretty much random allocation of multiple numbers and words as far as I can see. |
|
| |
Maybe this is the ultimate aim, after a few million iterations the AI stops playing and declares the entire game a complete waste of time. Suggests table-tennis instead. |
|
| |
On the other hand, AI would increase the overall IQ of the playing field and audience. |
|
| |
Such a desperately sad level of cricket hatred here. Clearly you all need to spend some time in the 8th of 7 Memorial Attitude Re-Alignment Centre.
You'll be pleased to know that I shall forego the oportunity to have a rant about so-called 'AI' and address the issues in the idea!
// This in itself is an iterative process.//
Only very broadly speaking, I would say. Once you go into detail you will find that sometimes information is lost during the process & also that new innovations are being constantly tried out. From experience, I can tell you that cricketers (and I realise that you are just using cricket as an example here) are constantly practising & trying out new ways of doing stuff. This is not always a good thing as the sad state of the England Test side amply demonstrates.
A better sporting example would be professional cycling. The influence of computer analysis on performance in that sport has been remarkable. In particular in time trialing, where performance has improved dramatically. However, this has brought with it serious concerns about the health of the sport & the riders.
Firstly, that the riders are now adopting such extreme aero-dynamic positions on the bike that there are concerns about the rising number of serious injuries that riders sustain.
Secondly, that the sport has become less interesting. Teams know exactly how much effort their riders can sustain & for how long. They all have wattage monitors on their bikes so they can see how much power they are generating. Successful breakaways by lesser riders, in particular for the grand tours, are not nearly as common as they used to be because the big teams in the main pack can just set a pace that is generally, but not always, guaranteed to close the breakaway down at the optimum moment in the race.
The use of team radios is also a hot topic. Although that falls outside the scope of this idea, each of these things tends to re-inforce or even multiply the effects of the other.
Anyway, in conclusion I am against the idea in principle but recognise that, in practice, the technology will get used regardless. So neither fish nor bun from me. |
|
| |
I too would like to see more cutting-edge computing resources devoted to cricket development. What else are we using it for? Curing diseases? Averting climate change? Finding people's lost car fobs? Boring. None of those are named after insects, either—where's the charm, entomologically? |
|
| |