Spaces:
Sleeping
Sleeping
File size: 12,529 Bytes
b14c6e3 c18a9d1 b14c6e3 c18a9d1 b14c6e3 c18a9d1 b14c6e3 c18a9d1 b14c6e3 c18a9d1 b14c6e3 c18a9d1 b14c6e3 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 | """
Unit Tests for Reward Calculation
Tests reward shaping logic and component breakdown.
"""
import pytest
from adaptive_alert_triage.models import Action, Alert, Reward
from rewards.reward import (
calculate_reward,
calculate_system_failure_penalty,
calculate_episode_bonus,
get_reward_range,
create_reward_summary,
REWARD_CRITICAL_HANDLED,
REWARD_FAILURE_PREVENTED,
REWARD_FALSE_POSITIVE_IGNORED,
PENALTY_MISSED_CRITICAL,
)
class TestRewardCalculation:
"""Test core reward calculation logic."""
def test_critical_investigated_reward(self):
"""Test reward for correctly investigating critical alert."""
alert = Alert(
id="alert_001",
visible_severity=0.85,
confidence=0.9,
alert_type="CPU",
age=1,
true_severity=0.90, # Critical
is_correlated=False,
)
action = Action(alert_id="alert_001", action_type="INVESTIGATE")
reward = calculate_reward(action, alert)
assert reward.value == REWARD_CRITICAL_HANDLED, \
f"Expected {REWARD_CRITICAL_HANDLED}, got {reward.value}"
assert reward.components["critical_handled"] == REWARD_CRITICAL_HANDLED
assert reward.info["action_correct"] is True
def test_critical_escalated_reward(self):
"""Test reward for escalating critical alert."""
alert = Alert(
id="alert_002",
visible_severity=0.9,
confidence=0.95,
alert_type="SECURITY",
age=2,
true_severity=0.95,
is_correlated=False,
)
action = Action(alert_id="alert_002", action_type="ESCALATE")
reward = calculate_reward(action, alert)
# Escalate gets 90% of investigate reward
expected = REWARD_CRITICAL_HANDLED * 0.9
assert abs(reward.value - expected) < 0.01, \
f"Expected ~{expected}, got {reward.value}"
assert reward.info["action_correct"] is True
def test_false_positive_ignored_reward(self):
"""Test reward for correctly ignoring false positive."""
alert = Alert(
id="alert_003",
visible_severity=0.3,
confidence=0.4,
alert_type="DISK",
age=0,
true_severity=0.15, # False positive
is_correlated=False,
)
action = Action(alert_id="alert_003", action_type="IGNORE")
reward = calculate_reward(action, alert)
assert reward.value == REWARD_FALSE_POSITIVE_IGNORED, \
f"Expected {REWARD_FALSE_POSITIVE_IGNORED}, got {reward.value}"
assert reward.components["false_positive_ignored"] == REWARD_FALSE_POSITIVE_IGNORED
assert reward.info["action_correct"] is True
def test_critical_ignored_penalty(self):
"""Test penalty for ignoring critical alert."""
alert = Alert(
id="alert_004",
visible_severity=0.7,
confidence=0.8,
alert_type="SECURITY",
age=2,
true_severity=0.95, # Critical
is_correlated=False,
)
action = Action(alert_id="alert_004", action_type="IGNORE")
reward = calculate_reward(action, alert)
assert reward.value == PENALTY_MISSED_CRITICAL, \
f"Expected {PENALTY_MISSED_CRITICAL}, got {reward.value}"
assert reward.components["missed_critical"] == PENALTY_MISSED_CRITICAL
assert reward.info["action_correct"] is False
def test_unnecessary_investigation_penalty(self):
"""Test penalty for investigating false positive."""
alert = Alert(
id="alert_005",
visible_severity=0.35,
confidence=0.45,
alert_type="NETWORK",
age=0,
true_severity=0.20, # False positive
is_correlated=False,
)
action = Action(alert_id="alert_005", action_type="INVESTIGATE")
reward = calculate_reward(action, alert)
assert reward.value < 0.0, "Should be negative for wasted resources"
assert reward.components["unnecessary_invest"] < 0.0
def test_correlated_alert_bonus(self):
"""Test bonus for handling correlated alerts."""
alert = Alert(
id="alert_006",
visible_severity=0.8,
confidence=0.85,
alert_type="CPU",
age=1,
true_severity=0.85,
is_correlated=True, # Correlated
)
action = Action(alert_id="alert_006", action_type="INVESTIGATE")
reward = calculate_reward(action, alert)
# Should get critical + failure prevention bonus
expected = REWARD_CRITICAL_HANDLED + REWARD_FAILURE_PREVENTED
assert reward.value == expected, \
f"Expected {expected}, got {reward.value}"
assert reward.components["failure_prevented"] == REWARD_FAILURE_PREVENTED
def test_medium_alert_handling(self):
"""Test reward for medium severity alerts."""
alert = Alert(
id="alert_007",
visible_severity=0.6,
confidence=0.7,
alert_type="MEMORY",
age=1,
true_severity=0.55, # Medium
is_correlated=False,
)
action = Action(alert_id="alert_007", action_type="INVESTIGATE")
reward = calculate_reward(action, alert)
# Medium alerts get scaled reward based on severity
assert 0.0 < reward.value < REWARD_CRITICAL_HANDLED, \
"Medium alert should get moderate positive reward"
assert "medium_handled" in reward.components
def test_delay_action_rewards(self):
"""Test rewards for DELAY action."""
# Delaying medium alert (acceptable)
alert_medium = Alert(
id="alert_008",
visible_severity=0.5,
confidence=0.6,
alert_type="DISK",
age=0,
true_severity=0.50,
is_correlated=False,
)
action_delay = Action(alert_id="alert_008", action_type="DELAY")
reward_medium = calculate_reward(action_delay, alert_medium, {"max_investigations": 3})
assert reward_medium.value >= 0.0, "Delaying medium alert should be acceptable"
# Delaying critical alert (risky)
alert_critical = Alert(
id="alert_009",
visible_severity=0.85,
confidence=0.9,
alert_type="CPU",
age=2,
true_severity=0.90,
is_correlated=False,
)
action_delay_crit = Action(alert_id="alert_009", action_type="DELAY")
reward_critical = calculate_reward(action_delay_crit, alert_critical)
assert reward_critical.value < 0.0, "Delaying critical alert should be penalized"
class TestRewardComponents:
"""Test reward component breakdown."""
def test_reward_has_components(self):
"""Test that all rewards include component breakdown."""
alert = Alert(
id="a1", visible_severity=0.9, confidence=0.9, alert_type="CPU",
age=1, true_severity=0.9
)
action = Action(alert_id="a1", action_type="INVESTIGATE")
reward = calculate_reward(action, alert)
assert isinstance(reward.components, dict)
assert len(reward.components) > 0
assert sum(reward.components.values()) == reward.value
def test_reward_info_fields(self):
"""Test that reward info contains useful debugging information."""
alert = Alert(
id="a1", visible_severity=0.9, confidence=0.9, alert_type="CPU",
age=1, true_severity=0.9
)
action = Action(alert_id="a1", action_type="INVESTIGATE")
reward = calculate_reward(action, alert)
assert "alert_id" in reward.info
assert "true_severity" in reward.info
assert "is_critical" in reward.info
assert "is_false_positive" in reward.info
assert "action_correct" in reward.info
class TestAuxiliaryFunctions:
"""Test auxiliary reward functions."""
def test_system_failure_penalty(self):
"""Test system failure penalty calculation."""
penalty_1 = calculate_system_failure_penalty(1)
penalty_3 = calculate_system_failure_penalty(3)
assert penalty_1 < 0.0
assert penalty_3 < penalty_1
def test_episode_bonus_high_accuracy(self):
"""Test episode bonus for high accuracy."""
bonus = calculate_episode_bonus(correct_actions=85, total_actions=100, failures_count=0)
assert bonus > 0.0, "High accuracy should give bonus"
def test_episode_bonus_perfect(self):
"""Test episode bonus for perfect performance."""
bonus_perfect = calculate_episode_bonus(
correct_actions=100, total_actions=100, failures_count=0
)
bonus_high = calculate_episode_bonus(
correct_actions=85, total_actions=100, failures_count=0
)
assert bonus_perfect > bonus_high, "Perfect should get higher bonus"
def test_episode_bonus_with_failures(self):
"""Test that failures reduce episode bonus."""
bonus_no_fail = calculate_episode_bonus(
correct_actions=80, total_actions=100, failures_count=0
)
bonus_with_fail = calculate_episode_bonus(
correct_actions=80, total_actions=100, failures_count=2
)
assert bonus_no_fail > bonus_with_fail, "Failures should reduce bonus"
def test_reward_range(self):
"""Test reward range calculation."""
min_r, max_r = get_reward_range()
assert min_r < 0.0, "Min reward should be negative (penalty)"
assert max_r > 0.0, "Max reward should be positive"
assert max_r >= abs(min_r) - 0.01, "Max reward magnitude should be similar or exceed penalty"
def test_reward_summary_empty(self):
"""Test reward summary with empty list."""
summary = create_reward_summary([])
assert summary["total_reward"] == 0.0
assert summary["num_steps"] == 0
def test_reward_summary_aggregation(self):
"""Test reward summary aggregates correctly."""
rewards = [
Reward(value=10.0, components={"critical_handled": 10.0},
info={"action_correct": True}),
Reward(value=3.0, components={"false_positive_ignored": 3.0},
info={"action_correct": True}),
Reward(value=-2.0, components={"unnecessary_investigation": -2.0},
info={"action_correct": False}),
]
summary = create_reward_summary(rewards)
assert summary["total_reward"] == 11.0
assert summary["mean_reward"] == 11.0 / 3
assert summary["num_steps"] == 3
assert summary["correct_actions"] == 2
assert summary["accuracy"] == 2/3
assert "critical_handled" in summary["components"]
class TestRewardConsistency:
"""Test consistency and edge cases."""
def test_same_input_same_reward(self):
"""Test deterministic reward calculation."""
alert = Alert(
id="a1", visible_severity=0.9, confidence=0.9, alert_type="CPU",
age=1, true_severity=0.9
)
action = Action(alert_id="a1", action_type="INVESTIGATE")
reward1 = calculate_reward(action, alert)
reward2 = calculate_reward(action, alert)
assert reward1.value == reward2.value
def test_all_action_types_covered(self):
"""Test that all action types produce rewards."""
alert = Alert(
id="a1", visible_severity=0.6, confidence=0.7, alert_type="CPU",
age=1, true_severity=0.6
)
action_types = ["INVESTIGATE", "IGNORE", "ESCALATE", "DELAY"]
for action_type in action_types:
action = Action(alert_id="a1", action_type=action_type)
reward = calculate_reward(action, alert)
assert isinstance(reward.value, float), \
f"Action {action_type} should return numeric reward"
if __name__ == "__main__":
pytest.main([__file__, "-v"])
|