vkatg/dag_remediation_traces

DAG Remediation Traces

Author: Venkata Krishna Azith Teja Ganti Part of the ExposureGuard PHI Re-identification Risk Ecosystem

Input/output execution traces for budget-constrained PHI remediation planning over multimodal clinical records. Each record pairs a patient risk profile with a complete DAG planning trace: which actions were selected, which dependency injections fired, the topological execution order, and the final residual risk and cost.

Use this to train or benchmark alternative remediation planners, study cost/risk tradeoffs, or evaluate DAG construction under budget constraints across text, image, audio, and EHR modalities.

Splits

Split	Config	Records	plan_ready	below_threshold
train	default	6,000	5,051 (84.2%)	949 (15.8%)
test	default	1,500	1,277 (85.1%)	223 (14.9%)
test	hard	1,000	1,000 (100%)	0

The hard config fixes input risk between 0.75 and 0.99 and caps budget at 0.60, creating a stress-test set where every record requires planning under tight constraints.

Plan quality

Numbers are shown for all records and for planned-only (records where the planner actually ran). The gap between the two reflects the 15.8% of default train records that were already below the risk threshold and received empty plans. The hard split has no below-threshold records so both columns are the same.

Metric	default all	default planned	hard all	hard planned
avg actions per plan	3.47	4.12	5.50	5.50
avg topological depth	3.47	4.12	5.50	5.50
max topological depth	10	10	10	10
avg residual risk	0.0420	0.0261	0.0429	0.0429
avg total cost	0.3093	0.3093	0.4678	0.4678
plans reaching residual = 0.0	74.0%	74.0%	66.4%	66.4%
avg risk reduction	96.5%	96.5%	94.8%	94.8%
median risk reduction	100.0%	100.0%	100.0%	100.0%
records with injected deps	33.6%	33.6%	67.9%	67.9%
unique execution sequences	38	38	--	--

The injected deps rate nearly doubles in the hard split (33.6% to 67.9%) -- tighter budgets under high risk force more complex dependency chains.

These metrics are self-reported by the generator. They describe how the planner performs against its own scoring logic, not against externally validated PHI risk reduction.

Input risk distribution (default train)

Metric	Value
mean	0.5191
stdev	0.2701
min / max	0.0502 / 0.9899

Input risk distribution (hard)

Metric	Value
mean	0.8692
min / max	0.7500 / 0.9898

Quick Start

from datasets import load_dataset

ds = load_dataset("vkatg/dag_remediation_traces")
train = ds["train"]
test  = ds["test"]

import json

with open("data/train.jsonl") as f:
    records = [json.loads(line) for line in f]

planned = [r for r in records if r["status"] == "plan_ready"]
print(len(planned), "planned records")
print("avg residual risk:", sum(r["estimated_residual_risk"] for r in planned) / len(planned))

Schema

Field	Type	Description
patient_id	string	Synthetic patient identifier (P-000000 format)
input.risk_score	float	Re-identification risk score (0.0 to 1.0)
input.retok_prob	float	Re-tokenization probability (influences retokenize_text scoring)
input.active_modalities	list	Active data modalities: text, image, audio, ehr
input.budget	float	Total cost budget for the remediation plan
plan	list	Ordered list of actions with priority, cost, risk_delta, deps
trace.selected_actions	list	Actions chosen by the scorer before dependency injection
trace.injected_deps	list	Dependencies automatically injected to satisfy action prerequisites
trace.topo_order	list	Final topological execution order
trace.score_breakdown	dict	Per-action priority scores
estimated_residual_risk	float	Risk remaining after all actions execute
total_cost	float	Sum of costs across all actions including injected deps
status	string	plan_ready or below_threshold

Action Catalog (14 actions)

Action	Cost	Risk Reduction	Modalities
mask_direct_id	0.05	0.35	text, image
generalize_dob	0.03	0.12	text, ehr
suppress_geo	0.04	0.10	text, ehr
redact_image_face	0.12	0.20	image
redact_image_tag	0.08	0.14	image
strip_audio_voice	0.15	0.18	audio
anon_audio_content	0.10	0.12	audio
drop_rare_code	0.06	0.09	ehr
k_anon_table	0.20	0.22	ehr, text
perturb_numerics	0.07	0.07	ehr
retokenize_text	0.09	0.11	text
cross_modal_unlink	0.18	0.28	all
federated_noise	0.25	0.15	all
audit_log_purge	0.02	0.03	all

Dependency graph

Some actions require others to run first:

• retokenize_text requires mask_direct_id
• k_anon_table requires generalize_dob, suppress_geo, drop_rare_code
• cross_modal_unlink requires redact_image_face, strip_audio_voice, mask_direct_id
• federated_noise requires k_anon_table, cross_modal_unlink
• audit_log_purge requires federated_noise
• redact_image_tag requires redact_image_face
• anon_audio_content requires strip_audio_voice

When a selected action has unmet dependencies, those dependencies are injected automatically. 33.6% of planned records include at least one injected dependency.

Action frequency (train)

Action	Occurrences
mask_direct_id	4,208
generalize_dob	4,175
suppress_geo	3,312
cross_modal_unlink	1,388
retokenize_text	1,156
redact_image_tag	950
audit_log_purge	352
redact_image_face	306
strip_audio_voice	233
drop_rare_code	189

Sample Record

{
  "patient_id": "P-000001",
  "input": {
    "risk_score": 0.9051,
    "retok_prob": 0.2147,
    "active_modalities": ["ehr"],
    "budget": 0.86
  },
  "plan": [
    {"action": "generalize_dob", "priority": 8.1366, "cost": 0.03, "risk_delta": 0.12, "deps": []},
    {"action": "suppress_geo",   "priority": 5.0854, "cost": 0.04, "risk_delta": 0.10, "deps": []},
    {"action": "cross_modal_unlink", "priority": 3.1642, "cost": 0.18, "risk_delta": 0.28, "deps": ["redact_image_face", "strip_audio_voice", "mask_direct_id"]}
  ],
  "trace": {
    "selected_actions": ["generalize_dob", "suppress_geo", "cross_modal_unlink", "drop_rare_code", "audit_log_purge", "k_anon_table"],
    "injected_deps": ["redact_image_face", "strip_audio_voice", "mask_direct_id", "federated_noise"],
    "topo_order": ["generalize_dob", "suppress_geo", "redact_image_face", "strip_audio_voice", "mask_direct_id", "cross_modal_unlink", "drop_rare_code", "federated_noise", "audit_log_purge", "k_anon_table"],
    "score_breakdown": {"generalize_dob": 8.1366, "suppress_geo": 5.0854, "cross_modal_unlink": 3.1642}
  },
  "estimated_residual_risk": 0.0,
  "total_cost": 1.1,
  "status": "plan_ready"
}

Planner Logic

The scorer ranks actions by a return-on-investment signal:

score = (reduces * modality_weight + retok_bonus) / base_cost * log(1 + risk * 5)

modality_weight is sampled per record (0.8 to 1.5) for a random subset of active modalities, reflecting variable sensitivity across data types. retok_bonus (+0.15) fires for retokenize_text when retok_prob > 0.55.

Selection adds actions greedily by score until the budget is hit or the risk reduction target is met (target = max(0, risk - 0.20), minimum spend 0.10). Dependency injection runs after selection to ensure topological validity.

Reproduce

python generate_dag_traces.py

The default train/test splits are deterministic at seed 7. The hard split uses seed 42 with risk sampled from [0.75, 0.99] and budget from [0.30, 0.60].

Related Models

Model	Role
vkatg/exposureguard-dagplanner	Model trained/evaluated on this dataset
vkatg/exposureguard-fedcrdt-distill	Produces risk_score and retok_prob inputs
vkatg/exposureguard-dcpg-encoder	Alternative source of risk_score

Related Datasets

• vkatg/streaming-phi-deidentification-benchmark
• vkatg/phi-audit-trace-benchmark

Citation

@misc{ganti2025exposureguard,
  title        = {ExposureGuard: Cross-Modal PHI Re-identification Risk Scoring with DCPG and Federated CRDT Distillation},
  author       = {Ganti, Venkata Krishna Azith Teja},
  year         = {2025},
  doi          = {10.5281/zenodo.18865882},
  howpublished = {\url{https://huggingface.co/vkatg}}
}

License

MIT. Fully synthetic data. Contains no real patient information.