All authors
Claude Skills by VectorSpaceLab
github.com/VectorSpaceLab6,028 skills13 installs7,823 views
- Diayn Recovery Evaluation HarnessAssemble validator-compatible reduced DIAYN recovery experiments that exercise fixed skills, discriminator rewards, and policy updates.Votes: 0GitHub stars: 247
- Diayn Skill Prior ConditioningBuild fixed-prior latent skill schedules and per-timestep conditioning records for DIAYN-style unsupervised rollouts.Votes: 0GitHub stars: 247
- Fusion Model TaxonomyClassify infrared-visible fusion approaches into explicit and implicit mechanism categories from textual method descriptions.Votes: 0GitHub stars: 247
- Fusion Quality MetricsCompute deterministic no-reference proxy metrics for fused infrared-visible images, including entropy, contrast, edge preservation, and stability.Votes: 0GitHub stars: 247
- Modality Feature FusionFuse paired infrared and visible arrays while preserving thermal salience and visible texture through controllable local/global weights.Votes: 0GitHub stars: 247
- Proxy Fusion RecoveryRun a bounded proxy recovery that compares explicit/implicit-style fusion and performs a small optimizer step on fusion weights.Votes: 0GitHub stars: 247
- Fb Greedy PolicyExtract no-planning greedy policies from FB forward embeddings and a projected reward vector.Votes: 0GitHub stars: 247
- Fb Occupancy FactorizationCreate and validate reduced forward-backward successor-occupancy factorizations from reward-free transition tables.Votes: 0GitHub stars: 247
- Reduced Fb RecoveryRun a bounded soft-mode reduced recovery experiment for forward-backward representations with executable evidence.Votes: 0GitHub stars: 247
- Reward ProjectionProject late-specified rewards into backward representation coordinates for immediate FB policy adaptation.Votes: 0GitHub stars: 247
- Opal Hierarchical Latent ControlRelabel downstream data with OPAL primitive latents and evaluate high-level latent control over temporally extended actions.Votes: 0GitHub stars: 247
- Opal Offline Segment ProtocolPrepare fixed-horizon offline state-action segments for OPAL primitive discovery and downstream latent relabeling.Votes: 0GitHub stars: 247
- Opal Primitive Autoencoding ObjectiveTrain or check an OPAL-style primitive autoencoding objective with reconstruction loss and KL-style prior matching.Votes: 0GitHub stars: 247
- Opal Recovery EvaluationValidate OPAL recovery evidence, mechanism checks, source boundaries, and proxy metric comparison.Votes: 0GitHub stars: 247
- Generalized Policy ImprovementBuild a transferred greedy policy by maximizing action values over a library of prior policies or successor-feature heads.Votes: 0GitHub stars: 247
- Sf Transfer RecoveryRun a bounded successor-features plus GPI transfer recovery experiment for changing linear-reward RL tasks.Votes: 0GitHub stars: 247
- Successor Feature ModelCompute and validate successor-feature reward decompositions for shared-dynamics reinforcement-learning tasks with changing linear rewards.Votes: 0GitHub stars: 247
- Gpi Action SelectionSelect USFA transfer actions by generalized policy improvement over candidate policy encodings.Votes: 0GitHub stars: 247
- Linear Reward Successor FeaturesCompute linear multitask rewards, successor-feature values, and vector Bellman targets for USFA-style reinforcement-learning tasks.Votes: 0GitHub stars: 247
- Tabular Usfa RecoveryRun a bounded Trip-MDP proxy experiment that validates USFA TD learning and GPI transfer mechanics.Votes: 0GitHub stars: 247
- Usfa Policy ConditioningRepresent USFA policies with reward-weight encodings z and compute greedy policy actions under those encodings.Votes: 0GitHub stars: 247
- Combat Priority ScorerRank simplified AutoAscend combat actions with symbolic safety, damage, and line-of-fire checks.Votes: 0GitHub stars: 247
- Interruptible Strategy ControllerSelect and interrupt prioritized symbolic NetHack strategies with explicit action queues.Votes: 0GitHub stars: 247
- Proxy Recovery EvaluatorValidate soft recovery traces by checking AutoAscend mechanism evidence and pass-rate metrics.Votes: 0GitHub stars: 247
- State Memory ModelBuild persistent symbolic NetHack state memory from compact observations for AutoAscend-style agents.Votes: 0GitHub stars: 247
- Survival Resource RulesApply AutoAscend nutrition and emergency resource rules to symbolic NetHack state memory.Votes: 0GitHub stars: 247
- Goal Effect InductionInduce separated goal and effect rules from behavioral cloning traces while logging training evidence.Votes: 0GitHub stars: 247
- Reactive Controller EvaluationEvaluate GRAIL-style reactive behavioral clones with metrics, predictions, and mechanism checks.Votes: 0GitHub stars: 247
- Trace Schema PreparationPrepare fixed-rate behavioral cloning trace examples for GRAIL-style reduced recovery experiments.Votes: 0GitHub stars: 247
- Ewc Fisher ImportanceEstimate and normalize diagonal Fisher importance vectors for Elastic Weight Consolidation from first-order gradients.Votes: 0GitHub stars: 247
- Ewc PenaltyCompute Elastic Weight Consolidation's Fisher-weighted quadratic penalty, gradient, and additive multi-task quadratic terms.Votes: 0GitHub stars: 247
- Ewc Recovery EvaluationRun a bounded EWC retention recovery comparison and emit validator-compatible result, trace, and mechanism-check artifacts.Votes: 0GitHub stars: 247
- Ewc Task ProtocolBuild deterministic sequential-task protocols for Elastic Weight Consolidation recovery experiments where old-task data is withheld after task switches.Votes: 0GitHub stars: 247
- Mas Continual Recovery EvalEvaluate MAS in a bounded sequential-learning recovery and report forgetting plus mechanism checks.Votes: 0GitHub stars: 247
- Mas Importance AdaptationAccumulate MAS importance across tasks and diagnose unlabeled subset adaptation.Votes: 0GitHub stars: 247
- Mas Regularized TrainingApply the MAS quadratic regularizer during later-task training and log parameter drift evidence.Votes: 0GitHub stars: 247
- Mas Unlabeled ImportanceEstimate Memory Aware Synapses parameter importance from unlabeled inputs using output-sensitivity gradients.Votes: 0GitHub stars: 247
- Action Task ProtocolApply NLE-style movement commands and task rewards for reduced staircase and score experiments.Votes: 0GitHub stars: 247
- Nle Action SpaceValidate NLE full or reduced action-space choices and invalid-action penalties.Votes: 0GitHub stars: 247
- Nle Reduced Learning UpdateRun a deterministic reduced learning update over NLE-style symbolic features and rewards.Votes: 0GitHub stars: 247
- Nle Symbolic ObservationNormalize and validate NetHack Learning Environment symbolic observations for recovery harnesses.Votes: 0GitHub stars: 247
- Nle Task RewardsCompute reduced NLE task rewards and clipping for symbolic recovery experiments.Votes: 0GitHub stars: 247
- Recovery Evaluation HarnessValidate reduced NLE recovery outputs for metric metadata, mechanism checks, and source-boundary safety.Votes: 0GitHub stars: 247
- Recurrent Impala ProxyRun a tiny recurrent policy proxy with an optimizer step that mirrors NLE baseline training evidence.Votes: 0GitHub stars: 247
- Rnd Exploration BonusCompute Random Network Distillation novelty bonuses and simple predictor updates for symbolic state features.Votes: 0GitHub stars: 247
- Symbolic Observation AdapterParse NLE-style symbolic terminal observations into validated feature records for lightweight recovery experiments.Votes: 0GitHub stars: 247
- Rnd Dual Return PpoCompute separate intrinsic and extrinsic discounted returns for RND-style PPO with dual value heads.Votes: 0GitHub stars: 247
- Rnd Intrinsic RewardCompute and test Random Network Distillation intrinsic rewards from predictor error against a fixed deterministic target feature map.Votes: 0GitHub stars: 247
- Rnd Normalization StreamsMaintain RND observation whitening/clipping and intrinsic reward scaling statistics.Votes: 0GitHub stars: 247
- Rnd Recovery HarnessRun a bounded soft-mode recovery experiment that validates core RND mechanism evidence without Atari-scale training.Votes: 0GitHub stars: 247