Implement more EAs

[aadit-n3rdy]

Modularise the EA component, and implement more evolutionary algorithms to be used for the asteroids.

Edit: EA component has been modularised, additional EAs may be implemented

[priyansh-narang2308]

/assign

[Devpool-soc]

Alright @priyansh-narang2308! Consider yourself officially assigned as now. If you are thinking, "Whoa, hold up, I didn't sign up for this," no worries, just yell "/unassign" and we can pretend this never happened. But if you are sticking around, just know your clock for this ticks out on: 2026-01-11. Don't screw it up :)

[priyansh-narang2308]

Hybrid Crossover Genetic Algorithm

I’m improving how asteroids learn by moving from asexual to sexual reproduction
peviously, the system cloned the single best asteroid and applied random mutation. This was wrking but learning was slow and limited.
the new approach allows offspring to inherit traits from two strong parents,leading to faster & stable learning.

Change: Crossover Logic, what i am thinking to imp,emennt

In neural_network.py, a new crossover_from_parents method added.

• each neural network weight is inherited from either Parent A or Parent B
• selection is random (50–50) per weight
• a small mutation is still applied to allow new behaviors to emerge

Fitness Selection Update

In game.py,the selection logic updated:

• Instead of choosing only the best asteroid
• The top two elite asteroids are selected as parents
• use this "elite pair as the breeding pool

why is this better

• faster convergence:to combine good traits is more efficient than relying only on mutatio
• better behavior: offspring can merge strengths from different strategies like movemnet or the accuracy.
• more challenging gameplay: enemies can. evolve in smarter, and less predictable ways

[aadit-n3rdy]

@priyansh-narang2308 this works but please modularise the EA component, so future contributors can just rewrite a single module rather than changing code across multiple places. This will also net you additional bounty points.

[priyansh-narang2308]

Hey @aadit-n3rdy Sure, will keep that in mind, will start implementing it

[aadit-n3rdy]

/bounty 200 @priyansh-narang2308

[Devpool-soc]

Another day, another coin. Way to get that bounty @priyansh-narang2308 you glorious keyboard-tapping, coffee-sipping, vibe-coding witcher!

[aadit-n3rdy]

@priyansh-narang2308 good work with the current framework, feel free to add more algorithms

[priyansh-narang2308]

Sure