Merge pull request #2135 from borglab/hybrid-improvements

gtsam/discrete/DiscreteBayesNet.cpp

@@ -71,16 +71,14 @@ DiscreteValues DiscreteBayesNet::sample(DiscreteValues result) const {
 
 /* ************************************************************************* */
 // The implementation is: build the entire joint into one factor and then prune.
-// TODO(Frank): This can be quite expensive *unless* the factors have already
+// NOTE(Frank): This can be quite expensive *unless* the factors have already
 // been pruned before. Another, possibly faster approach is branch and bound
 // search to find the K-best leaves and then create a single pruned conditional.
 DiscreteBayesNet DiscreteBayesNet::prune(
     size_t maxNrLeaves, const std::optional<double>& marginalThreshold,
     DiscreteValues* fixedValues) const {
   // Multiply into one big conditional. NOTE: possibly quite expensive.
-  DiscreteConditional joint;
+  DiscreteConditional joint = this->joint();
-  for (const DiscreteConditional::shared_ptr& conditional : *this)
-    joint = joint * (*conditional);
 
   // Prune the joint. NOTE: imperative and, again, possibly quite expensive.
   DiscreteConditional pruned = joint;
@@ -122,6 +120,15 @@ DiscreteBayesNet DiscreteBayesNet::prune(
   return result;
 }
 
+/* *********************************************************************** */
+DiscreteConditional DiscreteBayesNet::joint() const {
+  DiscreteConditional joint;
+  for (const DiscreteConditional::shared_ptr& conditional : *this)
+    joint = joint * (*conditional);
+
+  return joint;
+}
+
 /* *********************************************************************** */
 std::string DiscreteBayesNet::markdown(
     const KeyFormatter& keyFormatter,

gtsam/discrete/DiscreteBayesNet.h

@@ -136,6 +136,16 @@ class GTSAM_EXPORT DiscreteBayesNet: public BayesNet<DiscreteConditional> {
                            const std::optional<double>& marginalThreshold = {},
                            DiscreteValues* fixedValues = nullptr) const;
 
+    /**
+     * @brief Multiply all conditionals into one big joint conditional
+     * and return it.
+     *
+     * NOTE: possibly quite expensive.
+     *
+     * @return DiscreteConditional
+     */
+    DiscreteConditional joint() const;
+
     ///@}
     /// @name Wrapper support
     /// @{

gtsam/discrete/TableFactor.cpp

@@ -108,7 +108,9 @@ static Eigen::SparseVector<double> ComputeSparseTable(
    *
    */
   auto op = [&](const Assignment<Key>& assignment, double p) {
-    if (p > 0) {
+    // Check if greater than 1e-11 because we consider
+    // smaller than that as numerically 0
+    if (p > 1e-11) {
       // Get all the keys involved in this assignment
       KeySet assignmentKeys;
       for (auto&& [k, _] : assignment) {

gtsam/hybrid/HybridBayesNet.cpp

@@ -53,11 +53,11 @@ HybridBayesNet HybridBayesNet::prune(
 
   // Prune discrete Bayes net
   DiscreteValues fixed;
-  auto prunedBN = marginal.prune(maxNrLeaves, marginalThreshold, &fixed);
+  DiscreteBayesNet prunedBN =
+      marginal.prune(maxNrLeaves, marginalThreshold, &fixed);
 
   // Multiply into one big conditional. NOTE: possibly quite expensive.
-  DiscreteConditional pruned;
+  DiscreteConditional pruned = prunedBN.joint();
-  for (auto &&conditional : prunedBN) pruned = pruned * (*conditional);
 
   // Set the fixed values if requested.
   if (marginalThreshold && fixedValues) {

gtsam/hybrid/HybridSmoother.cpp

@@ -86,13 +86,28 @@ Ordering HybridSmoother::maybeComputeOrdering(
 }
 
 /* ************************************************************************* */
-void HybridSmoother::removeFixedValues(
+HybridGaussianFactorGraph HybridSmoother::removeFixedValues(
+    const HybridGaussianFactorGraph &graph,
     const HybridGaussianFactorGraph &newFactors) {
-  for (Key key : newFactors.discreteKeySet()) {
+  // Initialize graph
+  HybridGaussianFactorGraph updatedGraph(graph);
+
+  for (DiscreteKey dkey : newFactors.discreteKeys()) {
+    Key key = dkey.first;
     if (fixedValues_.find(key) != fixedValues_.end()) {
+      // Add corresponding discrete factor to reintroduce the information
+      std::vector<double> probabilities(
+          dkey.second, (1 - *marginalThreshold_) / dkey.second);
+      probabilities[fixedValues_[key]] = *marginalThreshold_;
+      DecisionTreeFactor dtf({dkey}, probabilities);
+      updatedGraph.push_back(dtf);
+
+      // Remove fixed value
       fixedValues_.erase(key);
     }
   }
+
+  return updatedGraph;
 }
 
 /* ************************************************************************* */
@@ -126,6 +141,11 @@ void HybridSmoother::update(const HybridNonlinearFactorGraph &newFactors,
             << std::endl;
 #endif
 
+  if (marginalThreshold_) {
+    // Remove fixed values for discrete keys which are introduced in newFactors
+    updatedGraph = removeFixedValues(updatedGraph, newFactors);
+  }
+
   Ordering ordering = this->maybeComputeOrdering(updatedGraph, given_ordering);
 
 #if GTSAM_HYBRID_TIMING
@@ -145,9 +165,6 @@ void HybridSmoother::update(const HybridNonlinearFactorGraph &newFactors,
   }
 #endif
 
-  // Remove fixed values for discrete keys which are introduced in newFactors
-  removeFixedValues(newFactors);
-
 #ifdef DEBUG_SMOOTHER
   // Print discrete keys in the bayesNetFragment:
   std::cout << "Discrete keys in bayesNetFragment: ";

gtsam/hybrid/HybridSmoother.h

@@ -145,8 +145,19 @@ class GTSAM_EXPORT HybridSmoother {
   Ordering maybeComputeOrdering(const HybridGaussianFactorGraph& updatedGraph,
                                 const std::optional<Ordering> givenOrdering);
 
-  /// Remove fixed discrete values for discrete keys introduced in `newFactors`.
+  /**
-  void removeFixedValues(const HybridGaussianFactorGraph& newFactors);
+   * @brief Remove fixed discrete values for discrete keys
+   * introduced in `newFactors`, and reintroduce discrete factors
+   * with marginalThreshold_ as the probability value.
+   *
+   * @param graph The factor graph with previous conditionals added in.
+   * @param newFactors The new factors added to the smoother,
+   * used to check if a fixed discrete value has been reintroduced.
+   * @return HybridGaussianFactorGraph
+   */
+  HybridGaussianFactorGraph removeFixedValues(
+      const HybridGaussianFactorGraph& graph,
+      const HybridGaussianFactorGraph& newFactors);
 };
 
 }  // namespace gtsam

gtsam/hybrid/hybrid.i

@@ -152,6 +152,7 @@ class HybridBayesNet {
   gtsam::HybridGaussianFactorGraph toFactorGraph(
       const gtsam::VectorValues& measurements) const;
 
+  gtsam::DiscreteBayesNet discreteMarginal() const;
   gtsam::GaussianBayesNet choose(const gtsam::DiscreteValues& assignment) const;
 
   gtsam::HybridValues optimize() const;