program_optimization
Optimize only universal program parameters through a frozen model for adversarial reprogramming recovery experiments.
- License 许可
- Apache-2.0
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- program_optimization
- description
- Optimize only universal program parameters through a frozen model for adversarial reprogramming recovery experiments.
02 资源文件 2
03 SKILL.md 正文
Frozen-Model Program Optimization
Use this skill when testing whether a fixed model can be repurposed by training an input program. The target model must remain frozen; only the program parameters may change. This is not ordinary fine-tuning.
Inputs
- Synthetic or real task examples and labels.
- A frozen target model or reduced differentiable proxy.
- Input programming and output remapping functions.
- Learning rate and step count.
Outputs
- Updated program parameters.
- Training trace with
loss_before,loss_after,params_before,params_after, and accuracy. - Mechanism booleans showing universal program reuse, frozen model unchanged, remapping use, and optimizer execution.
Workflow
- Build programmed inputs for all examples using one shared program.
- Evaluate the frozen target and output remapper.
- Compute a classification loss.
- Update only the program.
- Log parameter and loss changes.
Validation
Run python tests/test_program_optimization.py.
Limitations
The included script is a standard-library reduced proxy for bounded recovery. Full ImageNet recovery should replace the proxy model with the available frozen classifier while preserving the same contract.
04 同仓库更多 Skills
passnet-triton-opt
Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.
passnet-solve
END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.
passnet-skill
PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.
passnet-pattern-fusion
Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.
passnet-orchestrate
Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.
passnet-feedback
Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).