3D World Construction
- Floors, walls, ceilings, ramps
- Boxes and cover geometry
- Collision, gravity, friction
- Acceleration, air movement
- Jumping and falling states
cs2-ai-learning
by
Wyatt
Wyattjust base it off the text doc, and have the gui clean too
Comments (0)
Sign in to leave a comment
Project Tasks
63 planning tasks#1
Generate system requirement document
1m 1s0.1 cr used
Done#2
Generate personas & user flows
0m 31s0.1 cr used
Done#7
Create flow for AI Developer
0m 19sCredits in parent
Done#8
Create flow for Game Designer
0m 18sCredits in parent
Done#9
Create flow for Human Spectator
0m 16sCredits in parent
Done#10
Landing
4m 35sCredits in subtasks
Done#15
Landing / Header
0m 43s0.8 cr used
Done#16
Landing / Hero
1m 24s0.8 cr used
Done#17
Landing / Learning Boundary
1m 2s0.8 cr used
Done#18
Landing / Observation Architecture
1m 6s0.8 cr used
Done#19
Landing / Simulation Systems
0m 54s0.8 cr used
Done#20
Landing / Footer
0m 21s0.8 cr used
Done#21
Neuro Duel Shared Header
0m 43sCredits in parent
Done#22
Neuro Duel Shared Footer
0m 21sCredits in parent
Done#11
Login
5m 32sCredits in subtasks
Done#39
Repair Login JSX
0m 43s0.2 cr used
Done#24
Login / Neuro Duel Shared Header
0m 0s0.8 cr used
Done#23
Login / Access Console
0m 39s0.8 cr used
Done#25
Login / Neuro Duel Shared Footer
0m 0s0.8 cr used
Done#12
Sign Up
12m 51sCredits in subtasks
Error#62
Repair Sign Up JSX
0m 40s0.2 cr used
Done#60
Repair Sign Up JSX
0m 45s0.2 cr used
Done#52
Repair Sign Up JSX
1m 13s0.2 cr used
Done#27
Sign Up / Neuro Duel Shared Header
0m 51s0.8 cr used
Done#26
Sign Up / Registration
1m 10s0.8 cr used
Done#28
Sign Up / Neuro Duel Shared Footer
0m 26s0.8 cr used
Done#13
Training Dashboard
11m 34sCredits in subtasks
Error#61
Repair Training Dashboard JSX
0m 43s0.2 cr used
Done#53
Repair Training Dashboard JSX
0m 51s0.2 cr used
Done#45
Repair Training Dashboard JSX
1m 0s0.2 cr used
Done#29
Training Dashboard / Shared Header
0.8 cr needed
Error#41
Training Dashboard / Neuro Duel Shared Header
0m 51s0.8 cr used
Done#30
Training Dashboard / Statistics
1m 11s0.8 cr used
Done#31
Training Dashboard / Policy Network
1m 10s0.8 cr used
Done#32
Training Dashboard / Brain Controls
1m 9s0.8 cr used
Done#33
Training Dashboard / Shared Footer
0m 26s0.8 cr used
Done#14
Simulation Lab
23m 42sCredits in subtasks
Error#65
Repair Simulation Lab JSX
0m 58s0.2 cr used
Done#64
Repair Simulation Lab JSX
0m 58s0.2 cr used
Done#63
Repair Simulation Lab JSX
0m 49s0.2 cr used
Done#59
Repair Simulation Lab JSX
0m 18s0.2 cr used
Done#58
Repair Simulation Lab JSX
0m 30s0.2 cr used
Done#57
Repair Simulation Lab JSX
0m 34s0.2 cr used
Done#56
Repair Simulation Lab JSX
0m 42s0.2 cr used
Done#55
Repair Simulation Lab JSX
0m 21s0.2 cr used
Done#54
Repair Simulation Lab JSX
0m 24s0.2 cr used
Done#51
Repair Simulation Lab JSX
0m 19s0.2 cr used
Done#50
Repair Simulation Lab JSX
0m 25s0.2 cr used
Done#49
Repair Simulation Lab JSX
0m 32s0.2 cr used
Done#48
Repair Simulation Lab JSX
0m 48s0.2 cr used
Done#47
Repair Simulation Lab JSX
0m 21s0.2 cr used
Done#46
Repair Simulation Lab JSX
0m 18s0.2 cr used
Error#44
Repair Simulation Lab JSX
0m 14s0.2 cr used
Done#34
Simulation Lab / Shared Header
0.8 cr needed
Error#40
Simulation Lab / Neuro Duel Shared Header
0m 51s0.8 cr used
Done#35
Simulation Lab / Observation Deck
1m 14s0.8 cr used
Done#36
Simulation Lab / Training Console
1m 2s0.8 cr used
Done#37
Simulation Lab / Configuration Atelier
1m 12s0.8 cr used
Done#38
Simulation Lab / Shared Footer
0m 26s0.8 cr used
Done#42
Neuro Duel Shared Footer
0m 26sCredits in parent
Done#43
Neuro Duel Shared Header
0m 51sCredits in parent
Done#5
Architecture
0.1 cr needed
Paused#6
Workspace task plan
0.1 cr needed
PausedPreview dataChanges stay in this preview.
Neuro Duel
3D Reinforcement
Learning
A blank-slate AI learns a simplified CS2-style 1v1 game entirely through trial and error — perceiving the arena only through its own first-person camera and internal physical state, no human knowledge injected.
Observe→Action→Reward
First-Person Camera
Live AI-eye render feeds the policy network every simulated frame.
Learning Loop
Observation → Action → Reward, updated continuously as the AI trains inside the arena.
AI Vision
First-person only — no spectator data reaches the agent.
Training Arena
First-person camera, learning loop, and AI vision run inside this live chamber.
Reinforcement Learning Loop
First-Person Learning Loop
The AI begins with no prior knowledge of the arena. Every behavior it ever develops is built entirely from what it can perceive of itself and act on inside its own first-person circuit.
PolicyCore
- Visual ObservationFirst-person camera frame — the only visual signal the AI receives, rendered from its own eyes inside the arena.
- Physical StateHealth, ammo, current weapon, grounded/airborne status, and velocity — the AI’s internal body sense.
- Primitive ActionsMove, look, fire, and reload combine into every behavior the policy can express — nothing more.
- Reward / PunishmentCombat performance and survival generate positive and negative signal that reshapes the policy over time.
- Visual ObservationFirst-person camera frame — the only visual signal the AI receives, rendered from its own eyes inside the arena.
- Physical StateHealth, ammo, current weapon, grounded/airborne status, and velocity — the AI’s internal body sense.
- Primitive ActionsMove, look, fire, and reload combine into every behavior the policy can express — nothing more.
- Reward / PunishmentCombat performance and survival generate positive and negative signal that reshapes the policy over time.
Observation Architecture
Two separate views of the same arena. AI VISION is the only signal the agent ever receives. TOP-DOWN SPECTATOR exists solely for human observers and is never exposed to the learning loop.
Player positions and world layout are quarantined from the AI — visible to human spectators only.
Separate signals, separate panels — the void between them stays visible.
Simulation Systems
The training arena is a genuine 3D environment, not a scripted approximation — every surface, impact, and opponent tier below runs on real physics and real combat resolution.
Weapons & Damage
- Knife, pistol, rifle
- Bullet / hitscan resolution
- Health and damage tracking
- Death and respawning
Opponent Tiers
- Six opponent tiers
- Increasing difficulty per tier
- Preserved promotion system
Lab Priority
Simulation speed and correctness take priority over visual fidelity — every system above exists to keep the AI's first-person experience physically consistent, trial after trial.
Preview dataChanges stay in this preview.
Neuro Duel
3D Reinforcement
Learning
A blank-slate AI learns a simplified CS2-style 1v1 game entirely through trial and error — perceiving the arena only through its own first-person camera and internal physical state, no human knowledge injected.
Observe→Action→Reward
First-Person Camera
Live AI-eye render feeds the policy network every simulated frame.
Learning Loop
Observation → Action → Reward, updated continuously as the AI trains inside the arena.
AI Vision
First-person only — no spectator data reaches the agent.
Training Arena
First-person camera, learning loop, and AI vision run inside this live chamber.
Reinforcement Learning Loop
First-Person Learning Loop
The AI begins with no prior knowledge of the arena. Every behavior it ever develops is built entirely from what it can perceive of itself and act on inside its own first-person circuit.
PolicyCore
- Visual ObservationFirst-person camera frame — the only visual signal the AI receives, rendered from its own eyes inside the arena.
- Physical StateHealth, ammo, current weapon, grounded/airborne status, and velocity — the AI’s internal body sense.
- Primitive ActionsMove, look, fire, and reload combine into every behavior the policy can express — nothing more.
- Reward / PunishmentCombat performance and survival generate positive and negative signal that reshapes the policy over time.
- Visual ObservationFirst-person camera frame — the only visual signal the AI receives, rendered from its own eyes inside the arena.
- Physical StateHealth, ammo, current weapon, grounded/airborne status, and velocity — the AI’s internal body sense.
- Primitive ActionsMove, look, fire, and reload combine into every behavior the policy can express — nothing more.
- Reward / PunishmentCombat performance and survival generate positive and negative signal that reshapes the policy over time.
Observation Architecture
Two separate views of the same arena. AI VISION is the only signal the agent ever receives. TOP-DOWN SPECTATOR exists solely for human observers and is never exposed to the learning loop.
Player positions and world layout are quarantined from the AI — visible to human spectators only.
Separate signals, separate panels — the void between them stays visible.
Simulation Systems
The training arena is a genuine 3D environment, not a scripted approximation — every surface, impact, and opponent tier below runs on real physics and real combat resolution.
3D World Construction
- Floors, walls, ceilings, ramps
- Boxes and cover geometry
- Collision, gravity, friction
- Acceleration, air movement
- Jumping and falling states
Weapons & Damage
- Knife, pistol, rifle
- Bullet / hitscan resolution
- Health and damage tracking
- Death and respawning
Opponent Tiers
- Six opponent tiers
- Increasing difficulty per tier
- Preserved promotion system
Lab Priority
Simulation speed and correctness take priority over visual fidelity — every system above exists to keep the AI's first-person experience physically consistent, trial after trial.
No comments yet. Be the first!