agentic-qe 2.2.0 → 2.2.2

package/dist/learning/algorithms/MAMLMetaLearner.js

@@ -0,0 +1,532 @@
+"use strict";
+/**
+ * MAMLMetaLearner - Model-Agnostic Meta-Learning for QE Agents
+ *
+ * Implements MAML-style meta-learning that enables agents to "learn how to learn".
+ * Agents can quickly adapt to new testing domains with just 5-10 examples.
+ *
+ * Key Concepts:
+ * - Inner Loop: Fast adaptation to new task with few gradient steps (5-10 examples)
+ * - Outer Loop: Learn initialization parameters that enable fast adaptation
+ * - Meta-Learning: After seeing few examples of new test pattern, agent performs well
+ *
+ * Algorithm:
+ * 1. Initialize meta-parameters θ (Q-table initialization)
+ * 2. For each task Ti in task distribution:
+ *    a. Sample K examples from Ti (support set)
+ *    b. Adapt: θ'i = θ - α∇Loss(θ, support) [inner loop]
+ *    c. Evaluate on query set from Ti
+ * 3. Update meta-parameters: θ = θ - β∇Loss(θ', query) [outer loop]
+ * 4. Result: θ is optimized for fast adaptation to new tasks
+ *
+ * Use Cases:
+ * - New testing framework adoption (5-10 examples → proficient)
+ * - New project domain (few examples → effective testing strategy)
+ * - API testing → UI testing transfer learning
+ */
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.MAMLMetaLearner = void 0;
+exports.createDefaultMAMLConfig = createDefaultMAMLConfig;
+const AbstractRLLearner_1 = require("./AbstractRLLearner");
+/**
+ * Default MAML configuration
+ */
+const DEFAULT_MAML_CONFIG = {
+    // Base RL config
+    learningRate: 0.01, // Outer loop meta-learning rate (β)
+    discountFactor: 0.95,
+    explorationRate: 0.2, // Lower exploration for meta-learning
+    explorationDecay: 0.99,
+    minExplorationRate: 0.05,
+    useExperienceReplay: false, // MAML uses episodic learning
+    replayBufferSize: 1000,
+    batchSize: 10, // Task batch size
+    // MAML-specific
+    innerLearningRate: 0.1, // α - higher for fast adaptation
+    innerSteps: 5, // Few-shot learning (5 gradient steps)
+    metaLearningRate: 0.01, // β - meta-parameter update
+    minTaskExamples: 5, // Support set size (few-shot)
+    maxTaskExamples: 10,
+    taskBatchSize: 5, // Meta-train on 5 tasks per batch
+    firstOrderApproximation: true // Faster computation
+};
+/**
+ * MAMLMetaLearner - Model-Agnostic Meta-Learning for Fast Adaptation
+ *
+ * Learns an initialization of Q-values that enables rapid adaptation to new
+ * testing tasks with minimal examples (5-10 shots).
+ *
+ * Example:
+ * - Agent trained on Jest, Mocha, Jasmine unit testing
+ * - Sees 5-10 examples of Vitest tests
+ * - Immediately generates high-quality Vitest tests
+ *
+ * Meta-Learning Process:
+ * 1. Sample batch of tasks (different testing scenarios)
+ * 2. For each task:
+ *    - Adapt Q-table with support set (inner loop)
+ *    - Evaluate adapted Q-table on query set
+ * 3. Compute meta-gradient from all tasks
+ * 4. Update meta-parameters (Q-table initialization)
+ */
+class MAMLMetaLearner extends AbstractRLLearner_1.AbstractRLLearner {
+    constructor(config = {}) {
+        const fullConfig = { ...DEFAULT_MAML_CONFIG, ...config };
+        super(fullConfig);
+        this.mamlConfig = fullConfig;
+        // Meta-parameters (learned Q-table initialization)
+        this.metaQTable = new Map();
+        // Initialize with base Q-table
+        this.copyQTable(this.qTable, this.metaQTable);
+        // Task organization
+        this.taskBuffer = new Map();
+        this.metaEpisodes = [];
+        this.metaStepCount = 0;
+        this.logger.info('MAMLMetaLearner initialized for fast adaptation', {
+            innerLearningRate: fullConfig.innerLearningRate,
+            innerSteps: fullConfig.innerSteps,
+            metaLearningRate: fullConfig.metaLearningRate,
+            supportSetSize: fullConfig.minTaskExamples
+        });
+    }
+    /**
+     * Update Q-value using base algorithm (delegates to wrapped algorithm if available)
+     * For MAML, this is called during inner loop adaptation
+     */
+    update(experience, nextAction) {
+        if (this.baseAlgorithm) {
+            this.baseAlgorithm.update(experience, nextAction);
+        }
+        else {
+            // Default: Q-Learning update
+            this.qLearningUpdate(experience);
+        }
+        // Buffer experience by task type for meta-learning
+        this.bufferExperience(experience);
+        this.stepCount++;
+    }
+    /**
+     * Q-Learning update rule (default inner loop algorithm)
+     */
+    qLearningUpdate(experience) {
+        const stateKey = this.encodeState(experience.state);
+        const actionKey = this.encodeAction(experience.action);
+        const nextStateKey = this.encodeState(experience.nextState);
+        // Get current Q-value
+        const currentQ = this.getQValueRaw(stateKey, actionKey);
+        // Get max Q-value for next state
+        const nextStateActions = this.qTable.get(nextStateKey);
+        const maxNextQ = nextStateActions && nextStateActions.size > 0
+            ? Math.max(...Array.from(nextStateActions.values()).map(qv => qv.value))
+            : 0;
+        // Q-learning update
+        const tdTarget = experience.reward + this.config.discountFactor * maxNextQ;
+        const tdError = tdTarget - currentQ;
+        const newQ = currentQ + this.config.learningRate * tdError;
+        this.setQValue(stateKey, actionKey, newQ);
+    }
+    /**
+     * Buffer experience by task type for meta-learning
+     */
+    bufferExperience(experience) {
+        const taskType = experience.taskType;
+        if (!this.taskBuffer.has(taskType)) {
+            this.taskBuffer.set(taskType, []);
+        }
+        const buffer = this.taskBuffer.get(taskType);
+        buffer.push(experience);
+        // Limit buffer size per task type
+        const maxBufferSize = this.mamlConfig.maxTaskExamples * 10;
+        if (buffer.length > maxBufferSize) {
+            buffer.shift(); // Remove oldest
+        }
+    }
+    /**
+     * Perform meta-learning update (outer loop)
+     * Learns Q-table initialization that enables fast adaptation
+     *
+     * This should be called periodically after collecting enough task examples
+     */
+    async performMetaUpdate() {
+        // Sample batch of tasks from buffer
+        const tasks = this.sampleMetaTasks();
+        if (tasks.length < this.mamlConfig.taskBatchSize) {
+            this.logger.debug(`Not enough tasks for meta-update (${tasks.length}/${this.mamlConfig.taskBatchSize})`);
+            return null;
+        }
+        // Meta-gradients accumulator
+        const metaGradients = new Map();
+        let totalPreAdaptLoss = 0;
+        let totalPostAdaptLoss = 0;
+        // For each task in the batch
+        for (const task of tasks) {
+            // 1. Clone meta-parameters for this task
+            const taskQTable = this.cloneQTable(this.metaQTable);
+            // 2. Inner loop: Adapt to task using support set
+            const preAdaptLoss = this.evaluateLoss(taskQTable, task.supportSet);
+            for (let step = 0; step < this.mamlConfig.innerSteps; step++) {
+                this.adaptToTask(taskQTable, task.supportSet, this.mamlConfig.innerLearningRate);
+            }
+            // 3. Evaluate adapted parameters on query set
+            const postAdaptLoss = this.evaluateLoss(taskQTable, task.querySet);
+            totalPreAdaptLoss += preAdaptLoss;
+            totalPostAdaptLoss += postAdaptLoss;
+            // 4. Compute meta-gradients
+            if (this.mamlConfig.firstOrderApproximation) {
+                // First-order MAML: ignore second derivatives
+                this.computeFirstOrderGradients(metaGradients, taskQTable, task.querySet);
+            }
+            else {
+                // Full MAML: compute through inner loop
+                this.computeSecondOrderGradients(metaGradients, this.metaQTable, task);
+            }
+        }
+        // 5. Update meta-parameters using accumulated gradients
+        this.updateMetaParameters(metaGradients, tasks.length);
+        // 6. Copy meta-parameters to main Q-table
+        this.copyQTable(this.metaQTable, this.qTable);
+        this.metaStepCount++;
+        // Create meta-episode record
+        const metaEpisode = {
+            episodeId: `meta-${this.metaStepCount}-${Date.now()}`,
+            tasks,
+            preAdaptationLoss: totalPreAdaptLoss / tasks.length,
+            postAdaptationLoss: totalPostAdaptLoss / tasks.length,
+            metaLoss: totalPostAdaptLoss / tasks.length,
+            improvement: ((totalPreAdaptLoss - totalPostAdaptLoss) / totalPreAdaptLoss) * 100,
+            timestamp: new Date()
+        };
+        this.metaEpisodes.push(metaEpisode);
+        this.logger.info('Meta-update completed', {
+            episode: metaEpisode.episodeId,
+            tasks: tasks.length,
+            preAdaptLoss: metaEpisode.preAdaptationLoss.toFixed(4),
+            postAdaptLoss: metaEpisode.postAdaptationLoss.toFixed(4),
+            improvement: `${metaEpisode.improvement.toFixed(2)}%`
+        });
+        return metaEpisode;
+    }
+    /**
+     * Sample batch of meta-tasks from task buffer
+     * Each task contains support set (for adaptation) and query set (for evaluation)
+     */
+    sampleMetaTasks() {
+        const tasks = [];
+        const taskTypes = Array.from(this.taskBuffer.keys());
+        // Sample up to taskBatchSize tasks
+        const numTasks = Math.min(this.mamlConfig.taskBatchSize, taskTypes.length);
+        for (let i = 0; i < numTasks; i++) {
+            const taskType = taskTypes[i % taskTypes.length];
+            const experiences = this.taskBuffer.get(taskType) || [];
+            if (experiences.length < this.mamlConfig.minTaskExamples * 2) {
+                continue; // Not enough examples for support + query sets
+            }
+            // Shuffle experiences
+            const shuffled = [...experiences].sort(() => Math.random() - 0.5);
+            // Split into support and query sets
+            const supportSize = this.mamlConfig.minTaskExamples;
+            const supportSet = shuffled.slice(0, supportSize);
+            const querySet = shuffled.slice(supportSize, supportSize * 2);
+            tasks.push({
+                id: `task-${taskType}-${Date.now()}`,
+                taskType,
+                supportSet,
+                querySet
+            });
+        }
+        return tasks;
+    }
+    /**
+     * Adapt Q-table to a specific task using support set (inner loop)
+     */
+    adaptToTask(qTable, supportSet, learningRate) {
+        for (const experience of supportSet) {
+            const stateKey = this.encodeState(experience.state);
+            const actionKey = this.encodeAction(experience.action);
+            const nextStateKey = this.encodeState(experience.nextState);
+            // Get current Q-value from task-specific table
+            const currentQ = this.getQValueFromTable(qTable, stateKey, actionKey);
+            // Get max Q-value for next state
+            const nextStateActions = qTable.get(nextStateKey);
+            const maxNextQ = nextStateActions && nextStateActions.size > 0
+                ? Math.max(...Array.from(nextStateActions.values()).map(qv => qv.value))
+                : 0;
+            // Q-learning update with inner learning rate
+            const tdTarget = experience.reward + this.config.discountFactor * maxNextQ;
+            const tdError = tdTarget - currentQ;
+            const newQ = currentQ + learningRate * tdError;
+            // Update task-specific Q-table
+            this.setQValueInTable(qTable, stateKey, actionKey, newQ);
+        }
+    }
+    /**
+     * Evaluate loss (TD error) on a set of experiences
+     */
+    evaluateLoss(qTable, experiences) {
+        let totalLoss = 0;
+        for (const experience of experiences) {
+            const stateKey = this.encodeState(experience.state);
+            const actionKey = this.encodeAction(experience.action);
+            const nextStateKey = this.encodeState(experience.nextState);
+            const currentQ = this.getQValueFromTable(qTable, stateKey, actionKey);
+            const nextStateActions = qTable.get(nextStateKey);
+            const maxNextQ = nextStateActions && nextStateActions.size > 0
+                ? Math.max(...Array.from(nextStateActions.values()).map(qv => qv.value))
+                : 0;
+            const tdTarget = experience.reward + this.config.discountFactor * maxNextQ;
+            const tdError = tdTarget - currentQ;
+            totalLoss += tdError * tdError; // Squared TD error
+        }
+        return experiences.length > 0 ? totalLoss / experiences.length : 0;
+    }
+    /**
+     * Compute first-order meta-gradients (FOMAML)
+     * Faster approximation that ignores second-order derivatives
+     */
+    computeFirstOrderGradients(metaGradients, adaptedQTable, querySet) {
+        for (const experience of querySet) {
+            const stateKey = this.encodeState(experience.state);
+            const actionKey = this.encodeAction(experience.action);
+            const nextStateKey = this.encodeState(experience.nextState);
+            const currentQ = this.getQValueFromTable(adaptedQTable, stateKey, actionKey);
+            const nextStateActions = adaptedQTable.get(nextStateKey);
+            const maxNextQ = nextStateActions && nextStateActions.size > 0
+                ? Math.max(...Array.from(nextStateActions.values()).map(qv => qv.value))
+                : 0;
+            const tdTarget = experience.reward + this.config.discountFactor * maxNextQ;
+            const tdError = tdTarget - currentQ;
+            // Gradient: ∂Loss/∂Q = -2 * TD-error
+            const gradient = -2 * tdError;
+            // Accumulate gradient
+            if (!metaGradients.has(stateKey)) {
+                metaGradients.set(stateKey, new Map());
+            }
+            const stateGradients = metaGradients.get(stateKey);
+            const currentGradient = stateGradients.get(actionKey) || 0;
+            stateGradients.set(actionKey, currentGradient + gradient);
+        }
+    }
+    /**
+     * Compute second-order meta-gradients (Full MAML)
+     * More accurate but computationally expensive
+     */
+    computeSecondOrderGradients(metaGradients, metaQTable, task) {
+        // For simplicity, use first-order approximation
+        // Full second-order computation requires computing Hessians
+        const adaptedQTable = this.cloneQTable(metaQTable);
+        for (let step = 0; step < this.mamlConfig.innerSteps; step++) {
+            this.adaptToTask(adaptedQTable, task.supportSet, this.mamlConfig.innerLearningRate);
+        }
+        this.computeFirstOrderGradients(metaGradients, adaptedQTable, task.querySet);
+    }
+    /**
+     * Update meta-parameters using accumulated gradients
+     */
+    updateMetaParameters(metaGradients, numTasks) {
+        for (const [stateKey, stateGradients] of metaGradients.entries()) {
+            for (const [actionKey, gradient] of stateGradients.entries()) {
+                const currentQ = this.getQValueFromTable(this.metaQTable, stateKey, actionKey);
+                // Average gradient over tasks
+                const avgGradient = gradient / numTasks;
+                // Meta-gradient descent
+                const newQ = currentQ - this.mamlConfig.metaLearningRate * avgGradient;
+                this.setQValueInTable(this.metaQTable, stateKey, actionKey, newQ);
+            }
+        }
+    }
+    /**
+     * Fast adaptation to new task (few-shot learning)
+     * Given 5-10 examples, quickly adapt Q-table for new testing domain
+     *
+     * @param examples Few examples of new task (5-10)
+     * @returns Adapted Q-table
+     */
+    async fastAdapt(examples) {
+        if (examples.length < this.mamlConfig.minTaskExamples) {
+            this.logger.warn(`Few-shot adaptation requires at least ${this.mamlConfig.minTaskExamples} examples, got ${examples.length}`);
+        }
+        // Clone meta-parameters
+        const adaptedQTable = this.cloneQTable(this.metaQTable);
+        // Perform inner loop adaptation
+        for (let step = 0; step < this.mamlConfig.innerSteps; step++) {
+            this.adaptToTask(adaptedQTable, examples, this.mamlConfig.innerLearningRate);
+        }
+        this.logger.info(`Fast adaptation completed with ${examples.length} examples in ${this.mamlConfig.innerSteps} steps`);
+        return adaptedQTable;
+    }
+    /**
+     * Get Q-value from specific Q-table (helper)
+     */
+    getQValueFromTable(qTable, stateKey, actionKey) {
+        const stateActions = qTable.get(stateKey);
+        if (!stateActions)
+            return 0;
+        const qValue = stateActions.get(actionKey);
+        return qValue?.value ?? 0;
+    }
+    /**
+     * Set Q-value in specific Q-table (helper)
+     */
+    setQValueInTable(qTable, stateKey, actionKey, value) {
+        if (!qTable.has(stateKey)) {
+            qTable.set(stateKey, new Map());
+        }
+        const stateActions = qTable.get(stateKey);
+        const currentQValue = stateActions.get(actionKey);
+        stateActions.set(actionKey, {
+            state: stateKey,
+            action: actionKey,
+            value,
+            updateCount: (currentQValue?.updateCount ?? 0) + 1,
+            lastUpdated: Date.now()
+        });
+    }
+    /**
+     * Get Q-value (raw, without creating entry)
+     */
+    getQValueRaw(stateKey, actionKey) {
+        return this.getQValueFromTable(this.qTable, stateKey, actionKey);
+    }
+    /**
+     * Clone Q-table
+     */
+    cloneQTable(source) {
+        const cloned = new Map();
+        for (const [stateKey, stateActions] of source.entries()) {
+            const clonedActions = new Map();
+            for (const [actionKey, qValue] of stateActions.entries()) {
+                clonedActions.set(actionKey, { ...qValue });
+            }
+            cloned.set(stateKey, clonedActions);
+        }
+        return cloned;
+    }
+    /**
+     * Copy Q-table from source to destination
+     */
+    copyQTable(source, destination) {
+        destination.clear();
+        for (const [stateKey, stateActions] of source.entries()) {
+            const copiedActions = new Map();
+            for (const [actionKey, qValue] of stateActions.entries()) {
+                copiedActions.set(actionKey, { ...qValue });
+            }
+            destination.set(stateKey, copiedActions);
+        }
+    }
+    /**
+     * Get meta-learning statistics
+     */
+    getMetaStatistics() {
+        const recentEpisodes = this.metaEpisodes.slice(-10);
+        return {
+            metaSteps: this.metaStepCount,
+            metaEpisodes: this.metaEpisodes.length,
+            avgPreAdaptLoss: recentEpisodes.length > 0
+                ? recentEpisodes.reduce((sum, e) => sum + e.preAdaptationLoss, 0) / recentEpisodes.length
+                : 0,
+            avgPostAdaptLoss: recentEpisodes.length > 0
+                ? recentEpisodes.reduce((sum, e) => sum + e.postAdaptationLoss, 0) / recentEpisodes.length
+                : 0,
+            avgImprovement: recentEpisodes.length > 0
+                ? recentEpisodes.reduce((sum, e) => sum + e.improvement, 0) / recentEpisodes.length
+                : 0,
+            taskTypes: this.taskBuffer.size,
+            bufferedExperiences: Array.from(this.taskBuffer.values()).reduce((sum, arr) => sum + arr.length, 0)
+        };
+    }
+    /**
+     * Get meta-episodes history
+     */
+    getMetaEpisodes() {
+        return [...this.metaEpisodes];
+    }
+    /**
+     * Clear task buffer
+     */
+    clearTaskBuffer() {
+        this.taskBuffer.clear();
+    }
+    /**
+     * Get default exploration rate
+     */
+    getDefaultExplorationRate() {
+        return this.mamlConfig.explorationRate ?? 0.2;
+    }
+    /**
+     * Get algorithm name
+     */
+    getAlgorithmName() {
+        return 'MAML';
+    }
+    /**
+     * Override getStatistics to include meta-learning metrics
+     */
+    getStatistics() {
+        return {
+            ...super.getStatistics(),
+            maml: this.getMetaStatistics()
+        };
+    }
+    /**
+     * Export meta-learner state
+     */
+    export() {
+        const baseExport = super.export();
+        // Serialize metaQTable
+        const serializedMetaQTable = {};
+        for (const [state, actions] of this.metaQTable.entries()) {
+            serializedMetaQTable[state] = {};
+            for (const [action, qValue] of actions.entries()) {
+                serializedMetaQTable[state][action] = qValue;
+            }
+        }
+        // Serialize taskBuffer
+        const serializedTaskBuffer = {};
+        for (const [taskType, experiences] of this.taskBuffer.entries()) {
+            serializedTaskBuffer[taskType] = experiences;
+        }
+        return {
+            ...baseExport,
+            metaQTable: serializedMetaQTable,
+            taskBuffer: serializedTaskBuffer,
+            metaEpisodes: this.metaEpisodes,
+            metaStepCount: this.metaStepCount
+        };
+    }
+    /**
+     * Import meta-learner state
+     */
+    import(state) {
+        super.import(state);
+        // Deserialize metaQTable
+        this.metaQTable.clear();
+        for (const [stateKey, actions] of Object.entries(state.metaQTable)) {
+            const actionMap = new Map();
+            for (const [actionKey, qValue] of Object.entries(actions)) {
+                actionMap.set(actionKey, qValue);
+            }
+            this.metaQTable.set(stateKey, actionMap);
+        }
+        // Deserialize taskBuffer
+        this.taskBuffer.clear();
+        for (const [taskType, experiences] of Object.entries(state.taskBuffer)) {
+            this.taskBuffer.set(taskType, experiences);
+        }
+        this.metaEpisodes = state.metaEpisodes;
+        this.metaStepCount = state.metaStepCount;
+        this.logger.info('Imported MAML state', {
+            metaSteps: this.metaStepCount,
+            taskTypes: this.taskBuffer.size,
+            metaTableSize: this.metaQTable.size
+        });
+    }
+}
+exports.MAMLMetaLearner = MAMLMetaLearner;
+/**
+ * Create default MAML configuration
+ */
+function createDefaultMAMLConfig() {
+    return DEFAULT_MAML_CONFIG;
+}
+//# sourceMappingURL=MAMLMetaLearner.js.map

package/dist/learning/algorithms/MAMLMetaLearner.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"MAMLMetaLearner.js","sourceRoot":"","sources":["../../../src/learning/algorithms/MAMLMetaLearner.ts"],"names":[],"mappings":";AAAA;;;;;;;;;;;;;;;;;;;;;;;;GAwBG;;;AA0rBH,0DAEC;AA1rBD,2DAA0E;AA0B1E;;GAEG;AACH,MAAM,mBAAmB,GAAwB;IAC/C,iBAAiB;IACjB,YAAY,EAAE,IAAI,EAAE,oCAAoC;IACxD,cAAc,EAAE,IAAI;IACpB,eAAe,EAAE,GAAG,EAAE,sCAAsC;IAC5D,gBAAgB,EAAE,IAAI;IACtB,kBAAkB,EAAE,IAAI;IACxB,mBAAmB,EAAE,KAAK,EAAE,8BAA8B;IAC1D,gBAAgB,EAAE,IAAI;IACtB,SAAS,EAAE,EAAE,EAAE,kBAAkB;IAEjC,gBAAgB;IAChB,iBAAiB,EAAE,GAAG,EAAE,iCAAiC;IACzD,UAAU,EAAE,CAAC,EAAE,uCAAuC;IACtD,gBAAgB,EAAE,IAAI,EAAE,4BAA4B;IACpD,eAAe,EAAE,CAAC,EAAE,8BAA8B;IAClD,eAAe,EAAE,EAAE;IACnB,aAAa,EAAE,CAAC,EAAE,kCAAkC;IACpD,uBAAuB,EAAE,IAAI,CAAC,qBAAqB;CACpD,CAAC;AA0BF;;;;;;;;;;;;;;;;;;GAkBG;AACH,MAAa,eAAgB,SAAQ,qCAAiB;IAQpD,YAAY,SAA8B,EAAE;QAC1C,MAAM,UAAU,GAAG,EAAE,GAAG,mBAAmB,EAAE,GAAG,MAAM,EAAgB,CAAC;QACvE,KAAK,CAAC,UAAU,CAAC,CAAC;QAClB,IAAI,CAAC,UAAU,GAAG,UAAU,CAAC;QAE7B,mDAAmD;QACnD,IAAI,CAAC,UAAU,GAAG,IAAI,GAAG,EAAE,CAAC;QAE5B,+BAA+B;QAC/B,IAAI,CAAC,UAAU,CAAC,IAAI,CAAC,MAAM,EAAE,IAAI,CAAC,UAAU,CAAC,CAAC;QAE9C,oBAAoB;QACpB,IAAI,CAAC,UAAU,GAAG,IAAI,GAAG,EAAE,CAAC;QAC5B,IAAI,CAAC,YAAY,GAAG,EAAE,CAAC;QACvB,IAAI,CAAC,aAAa,GAAG,CAAC,CAAC;QAEvB,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,iDAAiD,EAAE;YAClE,iBAAiB,EAAE,UAAU,CAAC,iBAAiB;YAC/C,UAAU,EAAE,UAAU,CAAC,UAAU;YACjC,gBAAgB,EAAE,UAAU,CAAC,gBAAgB;YAC7C,cAAc,EAAE,UAAU,CAAC,eAAe;SAC3C,CAAC,CAAC;IACL,CAAC;IAED;;;OAGG;IACH,MAAM,CAAC,UAA0B,EAAE,UAAwB;QACzD,IAAI,IAAI,CAAC,aAAa,EAAE,CAAC;YACvB,IAAI,CAAC,aAAa,CAAC,MAAM,CAAC,UAAU,EAAE,UAAU,CAAC,CAAC;QACpD,CAAC;aAAM,CAAC;YACN,6BAA6B;YAC7B,IAAI,CAAC,eAAe,CAAC,UAAU,CAAC,CAAC;QACnC,CAAC;QAED,mDAAmD;QACnD,IAAI,CAAC,gBAAgB,CAAC,UAAU,CAAC,CAAC;QAElC,IAAI,CAAC,SAAS,EAAE,CAAC;IACnB,CAAC;IAED;;OAEG;IACK,eAAe,CAAC,UAA0B;QAChD,MAAM,QAAQ,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,KAAK,CAAC,CAAC;QACpD,MAAM,SAAS,GAAG,IAAI,CAAC,YAAY,CAAC,UAAU,CAAC,MAAM,CAAC,CAAC;QACvD,MAAM,YAAY,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,SAAS,CAAC,CAAC;QAE5D,sBAAsB;QACtB,MAAM,QAAQ,GAAG,IAAI,CAAC,YAAY,CAAC,QAAQ,EAAE,SAAS,CAAC,CAAC;QAExD,iCAAiC;QACjC,MAAM,gBAAgB,GAAG,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC;QACvD,MAAM,QAAQ,GAAG,gBAAgB,IAAI,gBAAgB,CAAC,IAAI,GAAG,CAAC;YAC5D,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,GAAG,KAAK,CAAC,IAAI,CAAC,gBAAgB,CAAC,MAAM,EAAE,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,KAAK,CAAC,CAAC;YACxE,CAAC,CAAC,CAAC,CAAC;QAEN,oBAAoB;QACpB,MAAM,QAAQ,GAAG,UAAU,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC,cAAc,GAAG,QAAQ,CAAC;QAC3E,MAAM,OAAO,GAAG,QAAQ,GAAG,QAAQ,CAAC;QACpC,MAAM,IAAI,GAAG,QAAQ,GAAG,IAAI,CAAC,MAAM,CAAC,YAAY,GAAG,OAAO,CAAC;QAE3D,IAAI,CAAC,SAAS,CAAC,QAAQ,EAAE,SAAS,EAAE,IAAI,CAAC,CAAC;IAC5C,CAAC;IAED;;OAEG;IACK,gBAAgB,CAAC,UAA0B;QACjD,MAAM,QAAQ,GAAG,UAAU,CAAC,QAAQ,CAAC;QAErC,IAAI,CAAC,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,QAAQ,CAAC,EAAE,CAAC;YACnC,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,QAAQ,EAAE,EAAE,CAAC,CAAC;QACpC,CAAC;QAED,MAAM,MAAM,GAAG,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,QAAQ,CAAE,CAAC;QAC9C,MAAM,CAAC,IAAI,CAAC,UAAU,CAAC,CAAC;QAExB,kCAAkC;QAClC,MAAM,aAAa,GAAG,IAAI,CAAC,UAAU,CAAC,eAAe,GAAG,EAAE,CAAC;QAC3D,IAAI,MAAM,CAAC,MAAM,GAAG,aAAa,EAAE,CAAC;YAClC,MAAM,CAAC,KAAK,EAAE,CAAC,CAAC,gBAAgB;QAClC,CAAC;IACH,CAAC;IAED;;;;;OAKG;IACH,KAAK,CAAC,iBAAiB;QACrB,oCAAoC;QACpC,MAAM,KAAK,GAAG,IAAI,CAAC,eAAe,EAAE,CAAC;QAErC,IAAI,KAAK,CAAC,MAAM,GAAG,IAAI,CAAC,UAAU,CAAC,aAAa,EAAE,CAAC;YACjD,IAAI,CAAC,MAAM,CAAC,KAAK,CAAC,qCAAqC,KAAK,CAAC,MAAM,IAAI,IAAI,CAAC,UAAU,CAAC,aAAa,GAAG,CAAC,CAAC;YACzG,OAAO,IAAI,CAAC;QACd,CAAC;QAED,6BAA6B;QAC7B,MAAM,aAAa,GAAG,IAAI,GAAG,EAA+B,CAAC;QAE7D,IAAI,iBAAiB,GAAG,CAAC,CAAC;QAC1B,IAAI,kBAAkB,GAAG,CAAC,CAAC;QAE3B,6BAA6B;QAC7B,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;YACzB,yCAAyC;YACzC,MAAM,UAAU,GAAG,IAAI,CAAC,WAAW,CAAC,IAAI,CAAC,UAAU,CAAC,CAAC;YAErD,iDAAiD;YACjD,MAAM,YAAY,GAAG,IAAI,CAAC,YAAY,CAAC,UAAU,EAAE,IAAI,CAAC,UAAU,CAAC,CAAC;YAEpE,KAAK,IAAI,IAAI,GAAG,CAAC,EAAE,IAAI,GAAG,IAAI,CAAC,UAAU,CAAC,UAAU,EAAE,IAAI,EAAE,EAAE,CAAC;gBAC7D,IAAI,CAAC,WAAW,CAAC,UAAU,EAAE,IAAI,CAAC,UAAU,EAAE,IAAI,CAAC,UAAU,CAAC,iBAAiB,CAAC,CAAC;YACnF,CAAC;YAED,8CAA8C;YAC9C,MAAM,aAAa,GAAG,IAAI,CAAC,YAAY,CAAC,UAAU,EAAE,IAAI,CAAC,QAAQ,CAAC,CAAC;YAEnE,iBAAiB,IAAI,YAAY,CAAC;YAClC,kBAAkB,IAAI,aAAa,CAAC;YAEpC,4BAA4B;YAC5B,IAAI,IAAI,CAAC,UAAU,CAAC,uBAAuB,EAAE,CAAC;gBAC5C,8CAA8C;gBAC9C,IAAI,CAAC,0BAA0B,CAAC,aAAa,EAAE,UAAU,EAAE,IAAI,CAAC,QAAQ,CAAC,CAAC;YAC5E,CAAC;iBAAM,CAAC;gBACN,wCAAwC;gBACxC,IAAI,CAAC,2BAA2B,CAAC,aAAa,EAAE,IAAI,CAAC,UAAU,EAAE,IAAI,CAAC,CAAC;YACzE,CAAC;QACH,CAAC;QAED,wDAAwD;QACxD,IAAI,CAAC,oBAAoB,CAAC,aAAa,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC;QAEvD,0CAA0C;QAC1C,IAAI,CAAC,UAAU,CAAC,IAAI,CAAC,UAAU,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC;QAE9C,IAAI,CAAC,aAAa,EAAE,CAAC;QAErB,6BAA6B;QAC7B,MAAM,WAAW,GAAgB;YAC/B,SAAS,EAAE,QAAQ,IAAI,CAAC,aAAa,IAAI,IAAI,CAAC,GAAG,EAAE,EAAE;YACrD,KAAK;YACL,iBAAiB,EAAE,iBAAiB,GAAG,KAAK,CAAC,MAAM;YACnD,kBAAkB,EAAE,kBAAkB,GAAG,KAAK,CAAC,MAAM;YACrD,QAAQ,EAAE,kBAAkB,GAAG,KAAK,CAAC,MAAM;YAC3C,WAAW,EAAE,CAAC,CAAC,iBAAiB,GAAG,kBAAkB,CAAC,GAAG,iBAAiB,CAAC,GAAG,GAAG;YACjF,SAAS,EAAE,IAAI,IAAI,EAAE;SACtB,CAAC;QAEF,IAAI,CAAC,YAAY,CAAC,IAAI,CAAC,WAAW,CAAC,CAAC;QAEpC,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,uBAAuB,EAAE;YACxC,OAAO,EAAE,WAAW,CAAC,SAAS;YAC9B,KAAK,EAAE,KAAK,CAAC,MAAM;YACnB,YAAY,EAAE,WAAW,CAAC,iBAAiB,CAAC,OAAO,CAAC,CAAC,CAAC;YACtD,aAAa,EAAE,WAAW,CAAC,kBAAkB,CAAC,OAAO,CAAC,CAAC,CAAC;YACxD,WAAW,EAAE,GAAG,WAAW,CAAC,WAAW,CAAC,OAAO,CAAC,CAAC,CAAC,GAAG;SACtD,CAAC,CAAC;QAEH,OAAO,WAAW,CAAC;IACrB,CAAC;IAED;;;OAGG;IACK,eAAe;QACrB,MAAM,KAAK,GAAe,EAAE,CAAC;QAC7B,MAAM,SAAS,GAAG,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,UAAU,CAAC,IAAI,EAAE,CAAC,CAAC;QAErD,mCAAmC;QACnC,MAAM,QAAQ,GAAG,IAAI,CAAC,GAAG,CAAC,IAAI,CAAC,UAAU,CAAC,aAAa,EAAE,SAAS,CAAC,MAAM,CAAC,CAAC;QAE3E,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,QAAQ,EAAE,CAAC,EAAE,EAAE,CAAC;YAClC,MAAM,QAAQ,GAAG,SAAS,CAAC,CAAC,GAAG,SAAS,CAAC,MAAM,CAAC,CAAC;YACjD,MAAM,WAAW,GAAG,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,QAAQ,CAAC,IAAI,EAAE,CAAC;YAExD,IAAI,WAAW,CAAC,MAAM,GAAG,IAAI,CAAC,UAAU,CAAC,eAAe,GAAG,CAAC,EAAE,CAAC;gBAC7D,SAAS,CAAC,+CAA+C;YAC3D,CAAC;YAED,sBAAsB;YACtB,MAAM,QAAQ,GAAG,CAAC,GAAG,WAAW,CAAC,CAAC,IAAI,CAAC,GAAG,EAAE,CAAC,IAAI,CAAC,MAAM,EAAE,GAAG,GAAG,CAAC,CAAC;YAElE,oCAAoC;YACpC,MAAM,WAAW,GAAG,IAAI,CAAC,UAAU,CAAC,eAAe,CAAC;YACpD,MAAM,UAAU,GAAG,QAAQ,CAAC,KAAK,CAAC,CAAC,EAAE,WAAW,CAAC,CAAC;YAClD,MAAM,QAAQ,GAAG,QAAQ,CAAC,KAAK,CAAC,WAAW,EAAE,WAAW,GAAG,CAAC,CAAC,CAAC;YAE9D,KAAK,CAAC,IAAI,CAAC;gBACT,EAAE,EAAE,QAAQ,QAAQ,IAAI,IAAI,CAAC,GAAG,EAAE,EAAE;gBACpC,QAAQ;gBACR,UAAU;gBACV,QAAQ;aACT,CAAC,CAAC;QACL,CAAC;QAED,OAAO,KAAK,CAAC;IACf,CAAC;IAED;;OAEG;IACK,WAAW,CACjB,MAAwC,EACxC,UAA4B,EAC5B,YAAoB;QAEpB,KAAK,MAAM,UAAU,IAAI,UAAU,EAAE,CAAC;YACpC,MAAM,QAAQ,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,KAAK,CAAC,CAAC;YACpD,MAAM,SAAS,GAAG,IAAI,CAAC,YAAY,CAAC,UAAU,CAAC,MAAM,CAAC,CAAC;YACvD,MAAM,YAAY,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,SAAS,CAAC,CAAC;YAE5D,+CAA+C;YAC/C,MAAM,QAAQ,GAAG,IAAI,CAAC,kBAAkB,CAAC,MAAM,EAAE,QAAQ,EAAE,SAAS,CAAC,CAAC;YAEtE,iCAAiC;YACjC,MAAM,gBAAgB,GAAG,MAAM,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC;YAClD,MAAM,QAAQ,GAAG,gBAAgB,IAAI,gBAAgB,CAAC,IAAI,GAAG,CAAC;gBAC5D,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,GAAG,KAAK,CAAC,IAAI,CAAC,gBAAgB,CAAC,MAAM,EAAE,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,KAAK,CAAC,CAAC;gBACxE,CAAC,CAAC,CAAC,CAAC;YAEN,6CAA6C;YAC7C,MAAM,QAAQ,GAAG,UAAU,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC,cAAc,GAAG,QAAQ,CAAC;YAC3E,MAAM,OAAO,GAAG,QAAQ,GAAG,QAAQ,CAAC;YACpC,MAAM,IAAI,GAAG,QAAQ,GAAG,YAAY,GAAG,OAAO,CAAC;YAE/C,+BAA+B;YAC/B,IAAI,CAAC,gBAAgB,CAAC,MAAM,EAAE,QAAQ,EAAE,SAAS,EAAE,IAAI,CAAC,CAAC;QAC3D,CAAC;IACH,CAAC;IAED;;OAEG;IACK,YAAY,CAClB,MAAwC,EACxC,WAA6B;QAE7B,IAAI,SAAS,GAAG,CAAC,CAAC;QAElB,KAAK,MAAM,UAAU,IAAI,WAAW,EAAE,CAAC;YACrC,MAAM,QAAQ,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,KAAK,CAAC,CAAC;YACpD,MAAM,SAAS,GAAG,IAAI,CAAC,YAAY,CAAC,UAAU,CAAC,MAAM,CAAC,CAAC;YACvD,MAAM,YAAY,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,SAAS,CAAC,CAAC;YAE5D,MAAM,QAAQ,GAAG,IAAI,CAAC,kBAAkB,CAAC,MAAM,EAAE,QAAQ,EAAE,SAAS,CAAC,CAAC;YAEtE,MAAM,gBAAgB,GAAG,MAAM,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC;YAClD,MAAM,QAAQ,GAAG,gBAAgB,IAAI,gBAAgB,CAAC,IAAI,GAAG,CAAC;gBAC5D,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,GAAG,KAAK,CAAC,IAAI,CAAC,gBAAgB,CAAC,MAAM,EAAE,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,KAAK,CAAC,CAAC;gBACxE,CAAC,CAAC,CAAC,CAAC;YAEN,MAAM,QAAQ,GAAG,UAAU,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC,cAAc,GAAG,QAAQ,CAAC;YAC3E,MAAM,OAAO,GAAG,QAAQ,GAAG,QAAQ,CAAC;YAEpC,SAAS,IAAI,OAAO,GAAG,OAAO,CAAC,CAAC,mBAAmB;QACrD,CAAC;QAED,OAAO,WAAW,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,SAAS,GAAG,WAAW,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,CAAC;IACrE,CAAC;IAED;;;OAGG;IACK,0BAA0B,CAChC,aAA+C,EAC/C,aAA+C,EAC/C,QAA0B;QAE1B,KAAK,MAAM,UAAU,IAAI,QAAQ,EAAE,CAAC;YAClC,MAAM,QAAQ,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,KAAK,CAAC,CAAC;YACpD,MAAM,SAAS,GAAG,IAAI,CAAC,YAAY,CAAC,UAAU,CAAC,MAAM,CAAC,CAAC;YACvD,MAAM,YAAY,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,SAAS,CAAC,CAAC;YAE5D,MAAM,QAAQ,GAAG,IAAI,CAAC,kBAAkB,CAAC,aAAa,EAAE,QAAQ,EAAE,SAAS,CAAC,CAAC;YAE7E,MAAM,gBAAgB,GAAG,aAAa,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC;YACzD,MAAM,QAAQ,GAAG,gBAAgB,IAAI,gBAAgB,CAAC,IAAI,GAAG,CAAC;gBAC5D,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,GAAG,KAAK,CAAC,IAAI,CAAC,gBAAgB,CAAC,MAAM,EAAE,CAAC,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,KAAK,CAAC,CAAC;gBACxE,CAAC,CAAC,CAAC,CAAC;YAEN,MAAM,QAAQ,GAAG,UAAU,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC,cAAc,GAAG,QAAQ,CAAC;YAC3E,MAAM,OAAO,GAAG,QAAQ,GAAG,QAAQ,CAAC;YAEpC,qCAAqC;YACrC,MAAM,QAAQ,GAAG,CAAC,CAAC,GAAG,OAAO,CAAC;YAE9B,sBAAsB;YACtB,IAAI,CAAC,aAAa,CAAC,GAAG,CAAC,QAAQ,CAAC,EAAE,CAAC;gBACjC,aAAa,CAAC,GAAG,CAAC,QAAQ,EAAE,IAAI,GAAG,EAAE,CAAC,CAAC;YACzC,CAAC;YACD,MAAM,cAAc,GAAG,aAAa,CAAC,GAAG,CAAC,QAAQ,CAAE,CAAC;YACpD,MAAM,eAAe,GAAG,cAAc,CAAC,GAAG,CAAC,SAAS,CAAC,IAAI,CAAC,CAAC;YAC3D,cAAc,CAAC,GAAG,CAAC,SAAS,EAAE,eAAe,GAAG,QAAQ,CAAC,CAAC;QAC5D,CAAC;IACH,CAAC;IAED;;;OAGG;IACK,2BAA2B,CACjC,aAA+C,EAC/C,UAA4C,EAC5C,IAAc;QAEd,gDAAgD;QAChD,4DAA4D;QAC5D,MAAM,aAAa,GAAG,IAAI,CAAC,WAAW,CAAC,UAAU,CAAC,CAAC;QAEnD,KAAK,IAAI,IAAI,GAAG,CAAC,EAAE,IAAI,GAAG,IAAI,CAAC,UAAU,CAAC,UAAU,EAAE,IAAI,EAAE,EAAE,CAAC;YAC7D,IAAI,CAAC,WAAW,CAAC,aAAa,EAAE,IAAI,CAAC,UAAU,EAAE,IAAI,CAAC,UAAU,CAAC,iBAAiB,CAAC,CAAC;QACtF,CAAC;QAED,IAAI,CAAC,0BAA0B,CAAC,aAAa,EAAE,aAAa,EAAE,IAAI,CAAC,QAAQ,CAAC,CAAC;IAC/E,CAAC;IAED;;OAEG;IACK,oBAAoB,CAC1B,aAA+C,EAC/C,QAAgB;QAEhB,KAAK,MAAM,CAAC,QAAQ,EAAE,cAAc,CAAC,IAAI,aAAa,CAAC,OAAO,EAAE,EAAE,CAAC;YACjE,KAAK,MAAM,CAAC,SAAS,EAAE,QAAQ,CAAC,IAAI,cAAc,CAAC,OAAO,EAAE,EAAE,CAAC;gBAC7D,MAAM,QAAQ,GAAG,IAAI,CAAC,kBAAkB,CAAC,IAAI,CAAC,UAAU,EAAE,QAAQ,EAAE,SAAS,CAAC,CAAC;gBAE/E,8BAA8B;gBAC9B,MAAM,WAAW,GAAG,QAAQ,GAAG,QAAQ,CAAC;gBAExC,wBAAwB;gBACxB,MAAM,IAAI,GAAG,QAAQ,GAAG,IAAI,CAAC,UAAU,CAAC,gBAAgB,GAAG,WAAW,CAAC;gBAEvE,IAAI,CAAC,gBAAgB,CAAC,IAAI,CAAC,UAAU,EAAE,QAAQ,EAAE,SAAS,EAAE,IAAI,CAAC,CAAC;YACpE,CAAC;QACH,CAAC;IACH,CAAC;IAED;;;;;;OAMG;IACH,KAAK,CAAC,SAAS,CAAC,QAA0B;QACxC,IAAI,QAAQ,CAAC,MAAM,GAAG,IAAI,CAAC,UAAU,CAAC,eAAe,EAAE,CAAC;YACtD,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,yCAAyC,IAAI,CAAC,UAAU,CAAC,eAAe,kBAAkB,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC;QAChI,CAAC;QAED,wBAAwB;QACxB,MAAM,aAAa,GAAG,IAAI,CAAC,WAAW,CAAC,IAAI,CAAC,UAAU,CAAC,CAAC;QAExD,gCAAgC;QAChC,KAAK,IAAI,IAAI,GAAG,CAAC,EAAE,IAAI,GAAG,IAAI,CAAC,UAAU,CAAC,UAAU,EAAE,IAAI,EAAE,EAAE,CAAC;YAC7D,IAAI,CAAC,WAAW,CAAC,aAAa,EAAE,QAAQ,EAAE,IAAI,CAAC,UAAU,CAAC,iBAAiB,CAAC,CAAC;QAC/E,CAAC;QAED,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,kCAAkC,QAAQ,CAAC,MAAM,gBAAgB,IAAI,CAAC,UAAU,CAAC,UAAU,QAAQ,CAAC,CAAC;QAEtH,OAAO,aAAa,CAAC;IACvB,CAAC;IAED;;OAEG;IACK,kBAAkB,CACxB,MAAwC,EACxC,QAAgB,EAChB,SAAiB;QAEjB,MAAM,YAAY,GAAG,MAAM,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;QAC1C,IAAI,CAAC,YAAY;YAAE,OAAO,CAAC,CAAC;QAE5B,MAAM,MAAM,GAAG,YAAY,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC;QAC3C,OAAO,MAAM,EAAE,KAAK,IAAI,CAAC,CAAC;IAC5B,CAAC;IAED;;OAEG;IACK,gBAAgB,CACtB,MAAwC,EACxC,QAAgB,EAChB,SAAiB,EACjB,KAAa;QAEb,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,QAAQ,CAAC,EAAE,CAAC;YAC1B,MAAM,CAAC,GAAG,CAAC,QAAQ,EAAE,IAAI,GAAG,EAAE,CAAC,CAAC;QAClC,CAAC;QACD,MAAM,YAAY,GAAG,MAAM,CAAC,GAAG,CAAC,QAAQ,CAAE,CAAC;QAE3C,MAAM,aAAa,GAAG,YAAY,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC;QAClD,YAAY,CAAC,GAAG,CAAC,SAAS,EAAE;YAC1B,KAAK,EAAE,QAAQ;YACf,MAAM,EAAE,SAAS;YACjB,KAAK;YACL,WAAW,EAAE,CAAC,aAAa,EAAE,WAAW,IAAI,CAAC,CAAC,GAAG,CAAC;YAClD,WAAW,EAAE,IAAI,CAAC,GAAG,EAAE;SACxB,CAAC,CAAC;IACL,CAAC;IAED;;OAEG;IACK,YAAY,CAAC,QAAgB,EAAE,SAAiB;QACtD,OAAO,IAAI,CAAC,kBAAkB,CAAC,IAAI,CAAC,MAAM,EAAE,QAAQ,EAAE,SAAS,CAAC,CAAC;IACnE,CAAC;IAED;;OAEG;IACK,WAAW,CAAC,MAAwC;QAC1D,MAAM,MAAM,GAAG,IAAI,GAAG,EAA+B,CAAC;QAEtD,KAAK,MAAM,CAAC,QAAQ,EAAE,YAAY,CAAC,IAAI,MAAM,CAAC,OAAO,EAAE,EAAE,CAAC;YACxD,MAAM,aAAa,GAAG,IAAI,GAAG,EAAkB,CAAC;YAChD,KAAK,MAAM,CAAC,SAAS,EAAE,MAAM,CAAC,IAAI,YAAY,CAAC,OAAO,EAAE,EAAE,CAAC;gBACzD,aAAa,CAAC,GAAG,CAAC,SAAS,EAAE,EAAE,GAAG,MAAM,EAAE,CAAC,CAAC;YAC9C,CAAC;YACD,MAAM,CAAC,GAAG,CAAC,QAAQ,EAAE,aAAa,CAAC,CAAC;QACtC,CAAC;QAED,OAAO,MAAM,CAAC;IAChB,CAAC;IAED;;OAEG;IACK,UAAU,CAChB,MAAwC,EACxC,WAA6C;QAE7C,WAAW,CAAC,KAAK,EAAE,CAAC;QAEpB,KAAK,MAAM,CAAC,QAAQ,EAAE,YAAY,CAAC,IAAI,MAAM,CAAC,OAAO,EAAE,EAAE,CAAC;YACxD,MAAM,aAAa,GAAG,IAAI,GAAG,EAAkB,CAAC;YAChD,KAAK,MAAM,CAAC,SAAS,EAAE,MAAM,CAAC,IAAI,YAAY,CAAC,OAAO,EAAE,EAAE,CAAC;gBACzD,aAAa,CAAC,GAAG,CAAC,SAAS,EAAE,EAAE,GAAG,MAAM,EAAE,CAAC,CAAC;YAC9C,CAAC;YACD,WAAW,CAAC,GAAG,CAAC,QAAQ,EAAE,aAAa,CAAC,CAAC;QAC3C,CAAC;IACH,CAAC;IAED;;OAEG;IACH,iBAAiB;QASf,MAAM,cAAc,GAAG,IAAI,CAAC,YAAY,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC;QAEpD,OAAO;YACL,SAAS,EAAE,IAAI,CAAC,aAAa;YAC7B,YAAY,EAAE,IAAI,CAAC,YAAY,CAAC,MAAM;YACtC,eAAe,EAAE,cAAc,CAAC,MAAM,GAAG,CAAC;gBACxC,CAAC,CAAC,cAAc,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,CAAC,EAAE,EAAE,CAAC,GAAG,GAAG,CAAC,CAAC,iBAAiB,EAAE,CAAC,CAAC,GAAG,cAAc,CAAC,MAAM;gBACzF,CAAC,CAAC,CAAC;YACL,gBAAgB,EAAE,cAAc,CAAC,MAAM,GAAG,CAAC;gBACzC,CAAC,CAAC,cAAc,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,CAAC,EAAE,EAAE,CAAC,GAAG,GAAG,CAAC,CAAC,kBAAkB,EAAE,CAAC,CAAC,GAAG,cAAc,CAAC,MAAM;gBAC1F,CAAC,CAAC,CAAC;YACL,cAAc,EAAE,cAAc,CAAC,MAAM,GAAG,CAAC;gBACvC,CAAC,CAAC,cAAc,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,CAAC,EAAE,EAAE,CAAC,GAAG,GAAG,CAAC,CAAC,WAAW,EAAE,CAAC,CAAC,GAAG,cAAc,CAAC,MAAM;gBACnF,CAAC,CAAC,CAAC;YACL,SAAS,EAAE,IAAI,CAAC,UAAU,CAAC,IAAI;YAC/B,mBAAmB,EAAE,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,UAAU,CAAC,MAAM,EAAE,CAAC,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,GAAG,EAAE,EAAE,CAAC,GAAG,GAAG,GAAG,CAAC,MAAM,EAAE,CAAC,CAAC;SACpG,CAAC;IACJ,CAAC;IAED;;OAEG;IACH,eAAe;QACb,OAAO,CAAC,GAAG,IAAI,CAAC,YAAY,CAAC,CAAC;IAChC,CAAC;IAED;;OAEG;IACH,eAAe;QACb,IAAI,CAAC,UAAU,CAAC,KAAK,EAAE,CAAC;IAC1B,CAAC;IAED;;OAEG;IACO,yBAAyB;QACjC,OAAO,IAAI,CAAC,UAAU,CAAC,eAAe,IAAI,GAAG,CAAC;IAChD,CAAC;IAED;;OAEG;IACH,gBAAgB;QACd,OAAO,MAAM,CAAC;IAChB,CAAC;IAED;;OAEG;IACH,aAAa;QAGX,OAAO;YACL,GAAG,KAAK,CAAC,aAAa,EAAE;YACxB,IAAI,EAAE,IAAI,CAAC,iBAAiB,EAAE;SAC/B,CAAC;IACJ,CAAC;IAED;;OAEG;IACH,MAAM;QAMJ,MAAM,UAAU,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC;QAElC,uBAAuB;QACvB,MAAM,oBAAoB,GAA2C,EAAE,CAAC;QACxE,KAAK,MAAM,CAAC,KAAK,EAAE,OAAO,CAAC,IAAI,IAAI,CAAC,UAAU,CAAC,OAAO,EAAE,EAAE,CAAC;YACzD,oBAAoB,CAAC,KAAK,CAAC,GAAG,EAAE,CAAC;YACjC,KAAK,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,IAAI,OAAO,CAAC,OAAO,EAAE,EAAE,CAAC;gBACjD,oBAAoB,CAAC,KAAK,CAAC,CAAC,MAAM,CAAC,GAAG,MAAM,CAAC;YAC/C,CAAC;QACH,CAAC;QAED,uBAAuB;QACvB,MAAM,oBAAoB,GAAqC,EAAE,CAAC;QAClE,KAAK,MAAM,CAAC,QAAQ,EAAE,WAAW,CAAC,IAAI,IAAI,CAAC,UAAU,CAAC,OAAO,EAAE,EAAE,CAAC;YAChE,oBAAoB,CAAC,QAAQ,CAAC,GAAG,WAAW,CAAC;QAC/C,CAAC;QAED,OAAO;YACL,GAAG,UAAU;YACb,UAAU,EAAE,oBAAoB;YAChC,UAAU,EAAE,oBAAoB;YAChC,YAAY,EAAE,IAAI,CAAC,YAAY;YAC/B,aAAa,EAAE,IAAI,CAAC,aAAa;SAClC,CAAC;IACJ,CAAC;IAED;;OAEG;IACH,MAAM,CAAC,KAA4C;QACjD,KAAK,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC;QAEpB,yBAAyB;QACzB,IAAI,CAAC,UAAU,CAAC,KAAK,EAAE,CAAC;QACxB,KAAK,MAAM,CAAC,QAAQ,EAAE,OAAO,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,KAAK,CAAC,UAAU,CAAC,EAAE,CAAC;YACnE,MAAM,SAAS,GAAG,IAAI,GAAG,EAAkB,CAAC;YAC5C,KAAK,MAAM,CAAC,SAAS,EAAE,MAAM,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,OAAO,CAAC,EAAE,CAAC;gBAC1D,SAAS,CAAC,GAAG,CAAC,SAAS,EAAE,MAAM,CAAC,CAAC;YACnC,CAAC;YACD,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,QAAQ,EAAE,SAAS,CAAC,CAAC;QAC3C,CAAC;QAED,yBAAyB;QACzB,IAAI,CAAC,UAAU,CAAC,KAAK,EAAE,CAAC;QACxB,KAAK,MAAM,CAAC,QAAQ,EAAE,WAAW,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,KAAK,CAAC,UAAU,CAAC,EAAE,CAAC;YACvE,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,QAAQ,EAAE,WAAW,CAAC,CAAC;QAC7C,CAAC;QAED,IAAI,CAAC,YAAY,GAAG,KAAK,CAAC,YAAY,CAAC;QACvC,IAAI,CAAC,aAAa,GAAG,KAAK,CAAC,aAAa,CAAC;QAEzC,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,qBAAqB,EAAE;YACtC,SAAS,EAAE,IAAI,CAAC,aAAa;YAC7B,SAAS,EAAE,IAAI,CAAC,UAAU,CAAC,IAAI;YAC/B,aAAa,EAAE,IAAI,CAAC,UAAU,CAAC,IAAI;SACpC,CAAC,CAAC;IACL,CAAC;CACF;AAtlBD,0CAslBC;AAED;;GAEG;AACH,SAAgB,uBAAuB;IACrC,OAAO,mBAAiC,CAAC;AAC3C,CAAC"}

package/dist/learning/algorithms/index.d.ts

@@ -7,21 +7,24 @@
  * - SARSALearner: On-policy TD(0) algorithm
  * - ActorCriticLearner: Advantage Actor-Critic (A2C) algorithm
  * - PPOLearner: Proximal Policy Optimization (PPO-Clip) algorithm
+ * - MAMLMetaLearner: Model-Agnostic Meta-Learning for fast adaptation
  */
 import { AbstractRLLearner, RLConfig, QValue } from './AbstractRLLearner';
 import { QLearning, QLearningConfig } from '../QLearning';
 import { SARSALearner, SARSAConfig } from './SARSALearner';
 import { ActorCriticLearner, ActorCriticConfig, createDefaultActorCriticConfig } from './ActorCriticLearner';
 import { PPOLearner, PPOConfig, createDefaultPPOConfig } from './PPOLearner';
+import { MAMLMetaLearner, MAMLConfig, createDefaultMAMLConfig } from './MAMLMetaLearner';
 export { AbstractRLLearner, RLConfig, QValue };
 export { QLearning, QLearningConfig };
 export { SARSALearner, SARSAConfig };
 export { ActorCriticLearner, ActorCriticConfig, createDefaultActorCriticConfig };
 export { PPOLearner, PPOConfig, createDefaultPPOConfig };
+export { MAMLMetaLearner, MAMLConfig, createDefaultMAMLConfig };
 /**
  * Supported RL algorithm types
  */
-export type RLAlgorithmType = 'q-learning' | 'sarsa' | 'actor-critic' | 'ppo';
+export type RLAlgorithmType = 'q-learning' | 'sarsa' | 'actor-critic' | 'ppo' | 'maml' | 'legacy';
 /**
  * Factory function to create RL algorithm instances
  */

package/dist/learning/algorithms/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/learning/algorithms/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAEH,OAAO,EAAE,iBAAiB,EAAE,QAAQ,EAAE,MAAM,EAAE,MAAM,qBAAqB,CAAC;AAC1E,OAAO,EAAE,SAAS,EAAE,eAAe,EAAE,MAAM,cAAc,CAAC;AAC1D,OAAO,EAAE,YAAY,EAAE,WAAW,EAAE,MAAM,gBAAgB,CAAC;AAC3D,OAAO,EAAE,kBAAkB,EAAE,iBAAiB,EAAE,8BAA8B,EAAE,MAAM,sBAAsB,CAAC;AAC7G,OAAO,EAAE,UAAU,EAAE,SAAS,EAAE,sBAAsB,EAAE,MAAM,cAAc,CAAC;AAE7E,OAAO,EAAE,iBAAiB,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC;AAC/C,OAAO,EAAE,SAAS,EAAE,eAAe,EAAE,CAAC;AACtC,OAAO,EAAE,YAAY,EAAE,WAAW,EAAE,CAAC;AACrC,OAAO,EAAE,kBAAkB,EAAE,iBAAiB,EAAE,8BAA8B,EAAE,CAAC;AACjF,OAAO,EAAE,UAAU,EAAE,SAAS,EAAE,sBAAsB,EAAE,CAAC;AAEzD;;GAEG;AACH,MAAM,MAAM,eAAe,GAAG,YAAY,GAAG,OAAO,GAAG,cAAc,GAAG,KAAK,CAAC;AAE9E;;GAEG;AACH,wBAAgB,iBAAiB,CAC/B,IAAI,EAAE,eAAe,EACrB,MAAM,CAAC,EAAE,GAAG,GACX,iBAAiB,CAanB"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/learning/algorithms/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAEH,OAAO,EAAE,iBAAiB,EAAE,QAAQ,EAAE,MAAM,EAAE,MAAM,qBAAqB,CAAC;AAC1E,OAAO,EAAE,SAAS,EAAE,eAAe,EAAE,MAAM,cAAc,CAAC;AAC1D,OAAO,EAAE,YAAY,EAAE,WAAW,EAAE,MAAM,gBAAgB,CAAC;AAC3D,OAAO,EAAE,kBAAkB,EAAE,iBAAiB,EAAE,8BAA8B,EAAE,MAAM,sBAAsB,CAAC;AAC7G,OAAO,EAAE,UAAU,EAAE,SAAS,EAAE,sBAAsB,EAAE,MAAM,cAAc,CAAC;AAC7E,OAAO,EAAE,eAAe,EAAE,UAAU,EAAE,uBAAuB,EAAE,MAAM,mBAAmB,CAAC;AAEzF,OAAO,EAAE,iBAAiB,EAAE,QAAQ,EAAE,MAAM,EAAE,CAAC;AAC/C,OAAO,EAAE,SAAS,EAAE,eAAe,EAAE,CAAC;AACtC,OAAO,EAAE,YAAY,EAAE,WAAW,EAAE,CAAC;AACrC,OAAO,EAAE,kBAAkB,EAAE,iBAAiB,EAAE,8BAA8B,EAAE,CAAC;AACjF,OAAO,EAAE,UAAU,EAAE,SAAS,EAAE,sBAAsB,EAAE,CAAC;AACzD,OAAO,EAAE,eAAe,EAAE,UAAU,EAAE,uBAAuB,EAAE,CAAC;AAEhE;;GAEG;AACH,MAAM,MAAM,eAAe,GAAG,YAAY,GAAG,OAAO,GAAG,cAAc,GAAG,KAAK,GAAG,MAAM,GAAG,QAAQ,CAAC;AAElG;;GAEG;AACH,wBAAgB,iBAAiB,CAC/B,IAAI,EAAE,eAAe,EACrB,MAAM,CAAC,EAAE,GAAG,GACX,iBAAiB,CAenB"}

package/dist/learning/algorithms/index.js

@@ -8,9 +8,10 @@
  * - SARSALearner: On-policy TD(0) algorithm
  * - ActorCriticLearner: Advantage Actor-Critic (A2C) algorithm
  * - PPOLearner: Proximal Policy Optimization (PPO-Clip) algorithm
+ * - MAMLMetaLearner: Model-Agnostic Meta-Learning for fast adaptation
  */
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.createDefaultPPOConfig = exports.PPOLearner = exports.createDefaultActorCriticConfig = exports.ActorCriticLearner = exports.SARSALearner = exports.QLearning = exports.AbstractRLLearner = void 0;
+exports.createDefaultMAMLConfig = exports.MAMLMetaLearner = exports.createDefaultPPOConfig = exports.PPOLearner = exports.createDefaultActorCriticConfig = exports.ActorCriticLearner = exports.SARSALearner = exports.QLearning = exports.AbstractRLLearner = void 0;
 exports.createRLAlgorithm = createRLAlgorithm;
 const AbstractRLLearner_1 = require("./AbstractRLLearner");
 Object.defineProperty(exports, "AbstractRLLearner", { enumerable: true, get: function () { return AbstractRLLearner_1.AbstractRLLearner; } });
@@ -24,6 +25,9 @@ Object.defineProperty(exports, "createDefaultActorCriticConfig", { enumerable: t
 const PPOLearner_1 = require("./PPOLearner");
 Object.defineProperty(exports, "PPOLearner", { enumerable: true, get: function () { return PPOLearner_1.PPOLearner; } });
 Object.defineProperty(exports, "createDefaultPPOConfig", { enumerable: true, get: function () { return PPOLearner_1.createDefaultPPOConfig; } });
+const MAMLMetaLearner_1 = require("./MAMLMetaLearner");
+Object.defineProperty(exports, "MAMLMetaLearner", { enumerable: true, get: function () { return MAMLMetaLearner_1.MAMLMetaLearner; } });
+Object.defineProperty(exports, "createDefaultMAMLConfig", { enumerable: true, get: function () { return MAMLMetaLearner_1.createDefaultMAMLConfig; } });
 /**
  * Factory function to create RL algorithm instances
  */
@@ -37,6 +41,8 @@ function createRLAlgorithm(type, config) {
             return new ActorCriticLearner_1.ActorCriticLearner(config ?? (0, ActorCriticLearner_1.createDefaultActorCriticConfig)());
         case 'ppo':
             return new PPOLearner_1.PPOLearner(config ?? (0, PPOLearner_1.createDefaultPPOConfig)());
+        case 'maml':
+            return new MAMLMetaLearner_1.MAMLMetaLearner(config ?? (0, MAMLMetaLearner_1.createDefaultMAMLConfig)());
         default:
             throw new Error(`Unknown RL algorithm type: ${type}`);
     }

package/dist/learning/algorithms/index.js.map

@@ -1 +1 @@
-{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../src/learning/algorithms/index.ts"],"names":[],"mappings":";AAAA;;;;;;;;;GASG;;;AAsBH,8CAgBC;AApCD,2DAA0E;AAMjE,kGANA,qCAAiB,OAMA;AAL1B,4CAA0D;AAMjD,0FANA,qBAAS,OAMA;AALlB,iDAA2D;AAMlD,6FANA,2BAAY,OAMA;AALrB,6DAA6G;AAMpG,mGANA,uCAAkB,OAMA;AAAqB,+GANA,mDAA8B,OAMA;AAL9E,6CAA6E;AAMpE,2FANA,uBAAU,OAMA;AAAa,uGANA,mCAAsB,OAMA;AAOtD;;GAEG;AACH,SAAgB,iBAAiB,CAC/B,IAAqB,EACrB,MAAY;IAEZ,QAAQ,IAAI,EAAE,CAAC;QACb,KAAK,YAAY;YACf,OAAO,IAAI,qBAAS,CAAC,MAAM,CAAC,CAAC;QAC/B,KAAK,OAAO;YACV,OAAO,IAAI,2BAAY,CAAC,MAAM,CAAC,CAAC;QAClC,KAAK,cAAc;YACjB,OAAO,IAAI,uCAAkB,CAAC,MAAM,IAAI,IAAA,mDAA8B,GAAE,CAAC,CAAC;QAC5E,KAAK,KAAK;YACR,OAAO,IAAI,uBAAU,CAAC,MAAM,IAAI,IAAA,mCAAsB,GAAE,CAAC,CAAC;QAC5D;YACE,MAAM,IAAI,KAAK,CAAC,8BAA8B,IAAI,EAAE,CAAC,CAAC;IAC1D,CAAC;AACH,CAAC"}
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../src/learning/algorithms/index.ts"],"names":[],"mappings":";AAAA;;;;;;;;;;GAUG;;;AAwBH,8CAkBC;AAxCD,2DAA0E;AAOjE,kGAPA,qCAAiB,OAOA;AAN1B,4CAA0D;AAOjD,0FAPA,qBAAS,OAOA;AANlB,iDAA2D;AAOlD,6FAPA,2BAAY,OAOA;AANrB,6DAA6G;AAOpG,mGAPA,uCAAkB,OAOA;AAAqB,+GAPA,mDAA8B,OAOA;AAN9E,6CAA6E;AAOpE,2FAPA,uBAAU,OAOA;AAAa,uGAPA,mCAAsB,OAOA;AANtD,uDAAyF;AAOhF,gGAPA,iCAAe,OAOA;AAAc,wGAPA,yCAAuB,OAOA;AAO7D;;GAEG;AACH,SAAgB,iBAAiB,CAC/B,IAAqB,EACrB,MAAY;IAEZ,QAAQ,IAAI,EAAE,CAAC;QACb,KAAK,YAAY;YACf,OAAO,IAAI,qBAAS,CAAC,MAAM,CAAC,CAAC;QAC/B,KAAK,OAAO;YACV,OAAO,IAAI,2BAAY,CAAC,MAAM,CAAC,CAAC;QAClC,KAAK,cAAc;YACjB,OAAO,IAAI,uCAAkB,CAAC,MAAM,IAAI,IAAA,mDAA8B,GAAE,CAAC,CAAC;QAC5E,KAAK,KAAK;YACR,OAAO,IAAI,uBAAU,CAAC,MAAM,IAAI,IAAA,mCAAsB,GAAE,CAAC,CAAC;QAC5D,KAAK,MAAM;YACT,OAAO,IAAI,iCAAe,CAAC,MAAM,IAAI,IAAA,yCAAuB,GAAE,CAAC,CAAC;QAClE;YACE,MAAM,IAAI,KAAK,CAAC,8BAA8B,IAAI,EAAE,CAAC,CAAC;IAC1D,CAAC;AACH,CAAC"}

package/dist/learning/index.d.ts

@@ -20,4 +20,12 @@ export * from './SwarmIntegration';
 export { FixRecommendationEngine } from './FixRecommendationEngine';
 export { AbstractRLLearner, RLConfig, QValue, SARSALearner, SARSAConfig, ActorCriticLearner, ActorCriticConfig, createDefaultActorCriticConfig, PPOLearner, PPOConfig, createDefaultPPOConfig, createRLAlgorithm } from './algorithms';
 export { ExperienceSharingProtocol, ExperienceSharingConfig, SharedExperience, SharingStats, PeerConnection, SharingEvent } from './ExperienceSharingProtocol';
+export { ExplainableLearning, ActionExplanation, ActionAlternative, ContributingExperience, DecisionFactor, StructuredExplanation, NaturalLanguageExplanation, DecisionType } from './ExplainableLearning';
+export { GossipPatternSharingProtocol, SharedPattern, PatternSharingConfig, PatternSharingStats, AntiEntropyResult, PatternSharingEvent } from './GossipPatternSharingProtocol';
+export { TransferLearningManager, TransferConfig, TransferMetrics, TransferMapping, FineTuningResult } from './TransferLearningManager';
+export type { QEDomain, DomainFeatures } from './TransferLearningManager';
+export { PerformanceOptimizer } from './PerformanceOptimizer';
+export type { PerformanceOptimizerConfig } from './PerformanceOptimizer';
+export { PrivacyManager, PrivacyLevel } from './PrivacyManager';
+export type { PrivacyConfig, SanitizedExperience, RetentionPolicyResult } from './PrivacyManager';
 //# sourceMappingURL=index.d.ts.map

package/dist/learning/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/learning/index.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,cAAc,SAAS,CAAC;AACxB,cAAc,kBAAkB,CAAC;AACjC,cAAc,aAAa,CAAC;AAC5B,cAAc,0BAA0B,CAAC;AACzC,cAAc,sBAAsB,CAAC;AACrC,cAAc,mBAAmB,CAAC;AAClC,cAAc,qBAAqB,CAAC;AACpC,cAAc,qBAAqB,CAAC;AACpC,cAAc,wBAAwB,CAAC;AACvC,cAAc,2BAA2B,CAAC;AAC1C,cAAc,uBAAuB,CAAC;AACtC,cAAc,oBAAoB,CAAC;AAGnC,OAAO,EACL,uBAAuB,EACxB,MAAM,2BAA2B,CAAC;AAInC,OAAO,EACL,iBAAiB,EACjB,QAAQ,EACR,MAAM,EACN,YAAY,EACZ,WAAW,EACX,kBAAkB,EAClB,iBAAiB,EACjB,8BAA8B,EAC9B,UAAU,EACV,SAAS,EACT,sBAAsB,EACtB,iBAAiB,EAClB,MAAM,cAAc,CAAC;AAGtB,OAAO,EACL,yBAAyB,EACzB,uBAAuB,EACvB,gBAAgB,EAChB,YAAY,EACZ,cAAc,EACd,YAAY,EACb,MAAM,6BAA6B,CAAC"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/learning/index.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,cAAc,SAAS,CAAC;AACxB,cAAc,kBAAkB,CAAC;AACjC,cAAc,aAAa,CAAC;AAC5B,cAAc,0BAA0B,CAAC;AACzC,cAAc,sBAAsB,CAAC;AACrC,cAAc,mBAAmB,CAAC;AAClC,cAAc,qBAAqB,CAAC;AACpC,cAAc,qBAAqB,CAAC;AACpC,cAAc,wBAAwB,CAAC;AACvC,cAAc,2BAA2B,CAAC;AAC1C,cAAc,uBAAuB,CAAC;AACtC,cAAc,oBAAoB,CAAC;AAGnC,OAAO,EACL,uBAAuB,EACxB,MAAM,2BAA2B,CAAC;AAInC,OAAO,EACL,iBAAiB,EACjB,QAAQ,EACR,MAAM,EACN,YAAY,EACZ,WAAW,EACX,kBAAkB,EAClB,iBAAiB,EACjB,8BAA8B,EAC9B,UAAU,EACV,SAAS,EACT,sBAAsB,EACtB,iBAAiB,EAClB,MAAM,cAAc,CAAC;AAGtB,OAAO,EACL,yBAAyB,EACzB,uBAAuB,EACvB,gBAAgB,EAChB,YAAY,EACZ,cAAc,EACd,YAAY,EACb,MAAM,6BAA6B,CAAC;AAGrC,OAAO,EACL,mBAAmB,EACnB,iBAAiB,EACjB,iBAAiB,EACjB,sBAAsB,EACtB,cAAc,EACd,qBAAqB,EACrB,0BAA0B,EAC1B,YAAY,EACb,MAAM,uBAAuB,CAAC;AAG/B,OAAO,EACL,4BAA4B,EAC5B,aAAa,EACb,oBAAoB,EACpB,mBAAmB,EACnB,iBAAiB,EACjB,mBAAmB,EACpB,MAAM,gCAAgC,CAAC;AAGxC,OAAO,EACL,uBAAuB,EACvB,cAAc,EACd,eAAe,EACf,eAAe,EACf,gBAAgB,EACjB,MAAM,2BAA2B,CAAC;AACnC,YAAY,EAAE,QAAQ,EAAE,cAAc,EAAE,MAAM,2BAA2B,CAAC;AAG1E,OAAO,EACL,oBAAoB,EACrB,MAAM,wBAAwB,CAAC;AAChC,YAAY,EAAE,0BAA0B,EAAE,MAAM,wBAAwB,CAAC;AAGzE,OAAO,EACL,cAAc,EACd,YAAY,EACb,MAAM,kBAAkB,CAAC;AAC1B,YAAY,EACV,aAAa,EACb,mBAAmB,EACnB,qBAAqB,EACtB,MAAM,kBAAkB,CAAC"}