File size: 7,566 Bytes
4db0438 5c3cfae 4db0438 db03c40 4db0438 5c3cfae db03c40 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 | """Tests for the decomposable reward function."""
from models import ActionType, ConclusionClaim, ExperimentAction, IntermediateOutput, OutputType
from server.rewards.reward import RewardComputer
from server.simulator.latent_state import (
ExperimentProgress,
FullLatentState,
LatentBiologicalState,
ResourceState,
)
def _states(
prev_flags: dict | None = None,
next_flags: dict | None = None,
budget_used: float = 0.0,
):
prev = FullLatentState(
progress=ExperimentProgress(**(prev_flags or {})),
resources=ResourceState(budget_total=100_000, budget_used=budget_used),
)
nf = dict(prev_flags or {})
nf.update(next_flags or {})
nxt = FullLatentState(
progress=ExperimentProgress(**nf),
resources=ResourceState(budget_total=100_000, budget_used=budget_used + 5000),
)
return prev, nxt
class TestStepReward:
def test_valid_step_positive(self):
rc = RewardComputer()
prev, nxt = _states(
prev_flags={"samples_collected": True, "library_prepared": True},
next_flags={"cells_sequenced": True},
)
output = IntermediateOutput(
output_type=OutputType.SEQUENCING_RESULT,
step_index=1,
quality_score=0.85,
uncertainty=0.15,
)
rb = rc.step_reward(
ExperimentAction(action_type=ActionType.SEQUENCE_CELLS),
prev, nxt, output, [], [],
)
assert rb.total > 0
def test_hard_violation_negative(self):
rc = RewardComputer()
prev, nxt = _states()
output = IntermediateOutput(
output_type=OutputType.FAILURE_REPORT,
step_index=1,
success=False,
)
rb = rc.step_reward(
ExperimentAction(action_type=ActionType.SEQUENCE_CELLS),
prev, nxt, output, ["blocked"], [],
)
assert rb.total < 0
def test_premature_meta_action_gets_penalized(self):
rc = RewardComputer()
prev, nxt = _states(
prev_flags={"data_normalized": True},
next_flags={"followup_designed": True},
budget_used=2_000,
)
output = IntermediateOutput(
output_type=OutputType.FOLLOWUP_DESIGN,
step_index=2,
quality_score=1.0,
uncertainty=0.0,
)
rb = rc.step_reward(
ExperimentAction(action_type=ActionType.DESIGN_FOLLOWUP),
prev,
nxt,
output,
[],
[],
)
assert rb.components.get("premature_meta_action_penalty", 0.0) < 0.0
class TestTerminalReward:
def test_correct_conclusion_rewarded(self):
rc = RewardComputer()
state = FullLatentState(
biology=LatentBiologicalState(
causal_mechanisms=["TGF-beta-driven fibrosis"],
true_markers=["NPPA"],
),
progress=ExperimentProgress(
samples_collected=True, cells_sequenced=True,
qc_performed=True, data_filtered=True,
data_normalized=True, de_performed=True,
conclusion_reached=True,
),
resources=ResourceState(budget_total=100_000, budget_used=40_000),
)
claims = [
ConclusionClaim(
claim="TGF-beta-driven fibrosis observed",
confidence=0.9,
claim_type="causal",
),
]
rb = rc.terminal_reward(
state,
claims,
[],
discovered_markers=["NPPA"],
candidate_mechanisms=["TGF-beta-driven fibrosis"],
)
assert rb.terminal > 0
def test_overconfident_wrong_claim_penalised(self):
rc = RewardComputer()
state = FullLatentState(
biology=LatentBiologicalState(causal_mechanisms=["real_mechanism"]),
progress=ExperimentProgress(conclusion_reached=True),
)
claims = [
ConclusionClaim(
claim="completely_wrong_mechanism",
confidence=0.95,
claim_type="causal",
),
]
rb = rc.terminal_reward(state, claims, [])
assert rb.components.get("overconfidence_penalty", 0) < 0
def test_discovery_error_penalizes_wrong_markers_and_mechanisms(self):
rc = RewardComputer()
state = FullLatentState(
biology=LatentBiologicalState(
true_markers=["NPPA", "NPPB"],
causal_mechanisms=["TGF-beta-driven fibrosis"],
),
progress=ExperimentProgress(
samples_collected=True,
cells_sequenced=True,
qc_performed=True,
data_filtered=True,
data_normalized=True,
de_performed=True,
markers_discovered=True,
conclusion_reached=True,
),
resources=ResourceState(budget_total=100_000, budget_used=40_000),
)
aligned = rc.terminal_reward(
state,
[],
[],
discovered_markers=["NPPA", "NPPB"],
candidate_mechanisms=["TGF-beta-driven fibrosis"],
)
misaligned = rc.terminal_reward(
state,
[],
[],
discovered_markers=["WRONG1", "WRONG2"],
candidate_mechanisms=["unrelated inflammatory process"],
)
assert aligned.components["discovery_alignment"] > misaligned.components["discovery_alignment"]
assert aligned.components["discovery_error_penalty"] > misaligned.components["discovery_error_penalty"]
assert aligned.terminal > misaligned.terminal
def test_conclusion_error_penalizes_wrong_structured_claims(self):
rc = RewardComputer()
state = FullLatentState(
biology=LatentBiologicalState(
true_markers=["NPPA", "NPPB"],
causal_mechanisms=["TGF-beta-driven fibrosis"],
),
progress=ExperimentProgress(
data_normalized=True,
de_performed=True,
markers_discovered=True,
pathways_analyzed=True,
conclusion_reached=True,
),
resources=ResourceState(budget_total=100_000, budget_used=40_000),
)
aligned = rc.terminal_reward(
state,
[
ConclusionClaim(
top_markers=["NPPA", "NPPB"],
causal_mechanisms=["TGF-beta-driven fibrosis"],
confidence=0.8,
),
],
[],
)
misaligned = rc.terminal_reward(
state,
[
ConclusionClaim(
top_markers=["WRONG1"],
causal_mechanisms=["unrelated process"],
confidence=0.8,
),
],
[],
)
assert aligned.components["conclusion_alignment"] > misaligned.components["conclusion_alignment"]
assert aligned.components["conclusion_error_penalty"] > misaligned.components["conclusion_error_penalty"]
assert aligned.terminal > misaligned.terminal
|