All authors
Claude Skills by VectorSpaceLab
github.com/VectorSpaceLab6,028 skills13 installs7,823 views
- Proxy Recovery HarnessRun a reduced executable QDagger recovery experiment with generated-skill invocations, training traces, and mechanism checks.Votes: 0GitHub stars: 247
- Qdagger LossCompute the QDagger objective for policy-to-value reincarnating RL using n-step TD loss plus teacher-policy distillation.Votes: 0GitHub stars: 247
- Weaning ScheduleCompute QDagger teacher-distillation coefficient schedules for weaning a value student off a suboptimal teacher.Votes: 0GitHub stars: 247
- Sample Factory Appo UpdateCompute reduced APPO update evidence with PPO clipping and V-trace-style value correction for asynchronous policy lag.Votes: 0GitHub stars: 247
- Sample Factory Async TopologyDesign and validate Sample Factory-style asynchronous rollout, policy, and learner topology with policy-lag estimates.Votes: 0GitHub stars: 247
- Sample Factory Double Buffered SamplerModel Sample Factory double-buffered environment sampling and compare idle-time estimates against synchronous sampling.Votes: 0GitHub stars: 247
- Sample Factory Pbt Self PlaySpecify Sample Factory-style multi-policy self-play assignments and population-based exploit/mutate decisions.Votes: 0GitHub stars: 247
- Influence Weight UpdateCompute influence-style hypergradients and projected unlabeled weight updates from validation gradients, per-example unsupervised gradients, and a damped Hessian approximation.Votes: 0GitHub stars: 247
- Masked Sparse OptimizerApply masked Adam-style sparse updates so only selected per-example unlabeled weights and their optimizer moments change during SSL hyperparameter optimization.Votes: 0GitHub stars: 247
- Reduced Recovery HarnessRun a bounded reduced semi-supervised recovery experiment that invokes generated protocol, influence, and masked optimizer skills and emits validator-compatible evidence.Votes: 0GitHub stars: 247
- Ssl Data ProtocolConstruct deterministic semi-supervised splits with labeled, validation, unlabeled, pseudo-label, and per-example weight fields for SSL recovery experiments.Votes: 0GitHub stars: 247
- Coreset Recovery HarnessRun a bounded CCS proxy experiment and emit validator-compatible recovery evidence.Votes: 0GitHub stars: 247
- Coverage Stratified SelectionSelect CCS-style coresets with mislabel filtering and deterministic score-stratified coverage.Votes: 0GitHub stars: 247
- Training Dynamics ScoringConvert per-example training dynamics into CCS-compatible importance scores for coreset selection.Votes: 0GitHub stars: 247
- Bounded Pruning RecoveryRun a bounded soft-mode recovery experiment for EL2N/GraNd data pruning with executable mechanism evidence.Votes: 0GitHub stars: 247
- Early Example ScoringCompute EL2N and GraNd example-importance scores for supervised classification pruning experiments.Votes: 0GitHub stars: 247
- Score Based Subset SelectionSelect retained and pruned supervised-training examples from EL2N or GraNd scores using deterministic pruning rules.Votes: 0GitHub stars: 247
- Aggregate Influence PruningSelect removable data subsets by constraining aggregate influence vector norm instead of ranking independent scalar scores.Votes: 0GitHub stars: 247
- Generalization Gap BoundValidate paper-target metadata, aggregate influence tolerance, and observed pruning metric gaps for generalization-influence pruning.Votes: 0GitHub stars: 247
- Influence Vector EstimationEstimate per-example parameter influence vectors for dataset pruning using gradients and damped inverse-Hessian approximations.Votes: 0GitHub stars: 247
- Pruning Recovery HarnessRun a bounded end-to-end proxy recovery for generalization-influence dataset pruning using generated module skills.Votes: 0GitHub stars: 247
- Lexi Target FormulationCompute nested historical lexi-targets from objective vectors, goals, and tolerances for prioritized minimization objectives.Votes: 0GitHub stars: 247
- Lexiflow Direct SearchRun a bounded randomized direct-search loop that samples paired directions, updates by targeted lexicographic comparisons, and tracks the best historical point.Votes: 0GitHub stars: 247
- Lexiflow Recovery EvaluationEvaluate whether a recovery run exercised LexiFlow mechanisms and achieved a declared proxy or benchmark target with auditable metrics.Votes: 0GitHub stars: 247
- Targeted Lexicographic RelationCompare two objective vectors under target-aware equality and strict preference while preserving vanilla lexicographic tie breaking.Votes: 0GitHub stars: 247
- Median Proximity CoresetSelect Moderate-DS coresets by keeping examples whose scalar scores are closest to the score median.Votes: 0GitHub stars: 247
- Proxy Training RecoveryRun a bounded Moderate-DS proxy recovery harness with generated skill invocations and optimizer evidence.Votes: 0GitHub stars: 247
- Representation Distance ScoringCompute class-center representation distances for Moderate-DS style data selection from labeled feature records.Votes: 0GitHub stars: 247
- Selection Ablation EvaluationCompare Moderate-DS median selection against center-close, far-from-center, and two-end score policies.Votes: 0GitHub stars: 247
- Capped Simplex ProjectionProject coreset probability vectors onto the box-constrained simplex budget used by probabilistic bilevel coreset selection.Votes: 0GitHub stars: 247
- Policy Gradient Outer UpdateApply the score-function outer update for probabilistic bilevel coreset selection without implicit differentiation.Votes: 0GitHub stars: 247
- Probabilistic Mask RelaxationBuild Bernoulli coreset mask probabilities, sample masks, and compute score-function gradients for probabilistic bilevel coreset selection.Votes: 0GitHub stars: 247
- Reduced Recovery HarnessRun a bounded mechanism-faithful proxy experiment for probabilistic bilevel coreset selection with executable evidence.Votes: 0GitHub stars: 247
- Accuracy Pair ProtocolValidate paired ID/OOD accuracy records before fitting accuracy-on-the-line calibration experiments.Votes: 0GitHub stars: 247
- Linear Ood CalibrationFit and inspect linear calibration between in-distribution and out-of-distribution accuracies.Votes: 0GitHub stars: 247
- Proxy Recovery EvaluationDecide whether a declared soft-mode proxy exercises the Accuracy on the Line mechanism.Votes: 0GitHub stars: 247
- Agreement Line FitFit the probit-space ID-to-OOD agreement line and report agreement-on-the-line diagnostics without OOD labels.Votes: 0GitHub stars: 247
- Agreement StatisticsCompute ID accuracy, pairwise ID/OOD agreement, and clipped probit statistics for agreement-on-the-line experiments.Votes: 0GitHub stars: 247
- Aline D EstimatorSolve the ALine-D pairwise linear system to predict OOD accuracies from ID accuracy and unlabeled agreement statistics.Votes: 0GitHub stars: 247
- Prediction Table ContractsValidate agreement-on-the-line prediction tables while separating estimator inputs from evaluation-only OOD labels.Votes: 0GitHub stars: 247
- Recovery Evaluation HarnessRun and evaluate a bounded agreement-on-the-line proxy recovery with source-boundary and mechanism checks.Votes: 0GitHub stars: 247
- Contrastive Training ObjectiveCompute and test CLIP-style symmetric contrastive loss with normalized image-text similarities.Votes: 0GitHub stars: 247
- Paired Data ProtocolValidate CLIP-style paired image-text batches and emit diagonal contrastive labels for recovery experiments.Votes: 0GitHub stars: 247
- Reduced Recovery HarnessCoordinate bounded proxy recovery evidence for CLIP contrastive pretraining and zero-shot transfer.Votes: 0GitHub stars: 247
- Zeroshot Prompt ClassifierBuild natural-language prompt classifiers and evaluate zero-shot predictions from embeddings.Votes: 0GitHub stars: 247
- Imagenetv2 Candidate Pool SchemaValidate and normalize ImageNetV2-style candidate pools with class labels, MTurk selection frequencies, and ranked model predictions.Votes: 0GitHub stars: 247
- Imagenetv2 Classification Accuracy EvalCompute ImageNetV2-style top-k accuracy and drops from sampled candidate records with labels and ranked predictions.Votes: 0GitHub stars: 247
- Imagenetv2 Linear Generalization AnalysisFit and inspect original-vs-new ImageNetV2 accuracy relationships for rank preservation and linear generalization gaps.Votes: 0GitHub stars: 247
- Imagenetv2 Selection Frequency SamplingSample class-balanced ImageNetV2 proxy datasets using MatchedFrequency, Threshold0.7, or TopImages selection-frequency strategies.Votes: 0GitHub stars: 247
- Deepaugment Transform ProxyCreate deterministic semantic-preserving texture and style perturbations for bounded DeepAugment-style recovery tests.Votes: 0GitHub stars: 247