Approximate furthest neighbor search tutorial

mlpack implements multiple strategies for approximate furthest neighbor search in its mlpack_approx_kfn and mlpack_kfn command-line programs (each program corresponds to different techniques). This tutorial discusses what problems these algorithms solve and how to use each of the techniques that mlpack implements.

Note that these functions are available as bindings to other languages too, and all the examples here can be adapted accordingly.

mlpack implements five approximate furthest neighbor search algorithms:

brute-force search (in mlpack_kfn)
single-tree search (in mlpack_kfn)
dual-tree search (in mlpack_kfn)
query-dependent approximate furthest neighbor (QDAFN) (in mlpack_approx_kfn)
DrusillaSelect (in mlpack_approx_kfn)

These methods are described in the following papers:

@inproceedings{curtin2013tree,
  title={Tree-Independent Dual-Tree Algorithms},
  author={Curtin, Ryan R. and March, William B. and Ram, Parikshit and Anderson,
      David V. and Gray, Alexander G. and Isbell Jr., Charles L.},
  booktitle={Proceedings of The 30th International Conference on Machine
      Learning (ICML '13)},
  pages={1435--1443},
  year={2013}
}

@incollection{pagh2015approximate,
  title={Approximate furthest neighbor in high dimensions},
  author={Pagh, Rasmus and Silvestri, Francesco and Sivertsen, Johan and Skala,
      Matthew},
  booktitle={Similarity Search and Applications},
  pages={3--14},
  year={2015},
  publisher={Springer}
}

@incollection{curtin2016fast,
  title={Fast approximate furthest neighbors with data-dependent candidate
      selection},
  author={Curtin, Ryan R., and Gardner, Andrew B.},
  booktitle={Similarity Search and Applications},
  pages={221--235},
  year={2016},
  publisher={Springer}
}

@article{curtin2018exploiting,
  title={Exploiting the structure of furthest neighbor search for fast
      approximate results},
  author={Curtin, Ryan R., and Echauz, Javier, and Gardner, Andrew B.},
  journal={Information Systems},
  year={2018},
  publisher={Elsevier}
}

The problem of furthest neighbor search is simple, and is the opposite of the much-more-studied nearest neighbor search problem. Given a set of reference points R (the set in which we are searching), and a set of query points Q (the set of points for which we want the furthest neighbor), our goal is to return the k furthest neighbors for each query point in Q:

k-argmax_{p_r in R} d(p_q, p_r).

In order to solve this problem, mlpack provides a number of interfaces.

two simple command-line executables to calculate approximate furthest neighbors
a simple C++ class for QDAFN
a simple C++ class for DrusillaSelect
a simple C++ class for tree-based and brute-force search

🔗 Which algorithm should be used?

There are three algorithms for furthest neighbor search that mlpack implements, and each is suited to a different setting. Below is some basic guidance on what should be used. Note that the question of “which algorithm should be used” is a very difficult question to answer, so the guidance below is just that—guidance—and may not be right for a particular problem.

DrusillaSelect is very fast and will perform extremely well for datasets with outliers or datasets with structure (like low-dimensional datasets embedded in high dimensions)
QDAFN is a random approach and therefore should be well-suited for datasets with little to no structure
The tree-based approaches (the KFN class and the mlpack_kfn program) is best suited for low-dimensional datasets, and is most effective when very small levels of approximation are desired, or when exact results are desired.
Dual-tree search is most useful when the query set is large and structured (like for all-furthest-neighbor search).
Single-tree search is more useful when the query set is small.

🔗 Command-line `mlpack_approx_kfn` and `mlpack_kfn`

mlpack provides two command-line programs to solve approximate furthest neighbor search:

mlpack_approx_kfn, for the QDAFN and DrusillaSelect approaches
mlpack_kfn, for exact and approximate tree-based approaches

These two programs allow a large number of algorithms to be used to find approximate furthest neighbors. Note that the mlpack_kfn program is also documented in the KNN tutorial page, as it shares options with the mlpack_knn program.

Below are several examples of how the mlpack_approx_kfn and mlpack_kfn programs might be used. The first examples focus on the mlpack_approx_kfn program, and the last few show how mlpack_kfn can be used to produce approximate results.

🔗 Calculate 5 furthest neighbors with default options

Here we have a query dataset queries.csv and a reference dataset refs.csv and we wish to find the 5 furthest neighbors of every query point in the reference dataset. We may do that with the mlpack_approx_kfn algorithm, using the default of the DrusillaSelect algorithm with default parameters.

$ mlpack_approx_kfn -q queries.csv -r refs.csv -v -k 5 -n n.csv -d d.csv
[INFO ] Loading 'refs.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Building DrusillaSelect model...
[INFO ] Model built.
[INFO ] Loading 'queries.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Searching for 5 furthest neighbors with DrusillaSelect...
[INFO ] Search complete.
[INFO ] Saving CSV data to 'n.csv'.
[INFO ] Saving CSV data to 'd.csv'.
[INFO ]
[INFO ] Execution parameters:
[INFO ]   algorithm: ds
[INFO ]   calculate_error: false
[INFO ]   distances_file: d.csv
[INFO ]   exact_distances_file: ""
[INFO ]   help: false
[INFO ]   info: ""
[INFO ]   input_model_file: ""
[INFO ]   k: 5
[INFO ]   neighbors_file: n.csv
[INFO ]   num_projections: 5
[INFO ]   num_tables: 5
[INFO ]   output_model_file: ""
[INFO ]   query_file: queries.csv
[INFO ]   reference_file: refs.csv
[INFO ]   verbose: true
[INFO ]   version: false
[INFO ]
[INFO ] Program timers:
[INFO ]   drusilla_select_construct: 0.000342s
[INFO ]   drusilla_select_search: 0.000780s
[INFO ]   loading_data: 0.010689s
[INFO ]   saving_data: 0.005585s
[INFO ]   total_time: 0.018592s

Convenient timers for parts of the program operation are printed. The results, saved in n.csv and d.csv, indicate the furthest neighbors and distances for each query point. The row of the output file indicates the query point that the results are for. The neighbors are listed from furthest to nearest; so, the 4th element in the 3rd row of d.csv indicates the distance between the 3rd query point in queries.csv and its approximate 4th furthest neighbor. Similarly, the same element in n.csv indicates the index of the approximate 4th furthest neighbor (with respect to refs.csv).

🔗 Specifying algorithm parameters for `DrusillaSelect`

The -p (--num_projections) and -t (--num_tables) parameters affect the running of the DrusillaSelect algorithm and the QDAFN algorithm. Specifically, larger values for each of these parameters will search more possible candidate furthest neighbors and produce better results (at the cost of runtime). More details on how each of these parameters works is available in the original papers, the mlpack source, or the documentation given by --help.

In the example below, we run DrusillaSelect to find 4 furthest neighbors using 10 tables and 2 points in each table. In this case we have chosen to omit the -n n.csv option, meaning that only the output candidate distances will be written to d.csv.

$ mlpack_approx_kfn -q queries.csv -r refs.csv -v -k 4 -n n.csv -d d.csv -t 10 -p 2
[INFO ] Loading 'refs.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Building DrusillaSelect model...
[INFO ] Model built.
[INFO ] Loading 'queries.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Searching for 4 furthest neighbors with DrusillaSelect...
[INFO ] Search complete.
[INFO ] Saving CSV data to 'n.csv'.
[INFO ] Saving CSV data to 'd.csv'.
[INFO ]
[INFO ] Execution parameters:
[INFO ]   algorithm: ds
[INFO ]   calculate_error: false
[INFO ]   distances_file: d.csv
[INFO ]   exact_distances_file: ""
[INFO ]   help: false
[INFO ]   info: ""
[INFO ]   input_model_file: ""
[INFO ]   k: 4
[INFO ]   neighbors_file: n.csv
[INFO ]   num_projections: 2
[INFO ]   num_tables: 10
[INFO ]   output_model_file: ""
[INFO ]   query_file: queries.csv
[INFO ]   reference_file: refs.csv
[INFO ]   verbose: true
[INFO ]   version: false
[INFO ]
[INFO ] Program timers:
[INFO ]   drusilla_select_construct: 0.000645s
[INFO ]   drusilla_select_search: 0.000551s
[INFO ]   loading_data: 0.008518s
[INFO ]   saving_data: 0.003734s
[INFO ]   total_time: 0.014019s

🔗 Using QDAFN instead of `DrusillaSelect`

The algorithm to be used for approximate furthest neighbor search can be specified with the --algorithm (-a) option to the mlpack_approx_kfn program. Below, we use the QDAFN algorithm instead of the default. We leave the -p and -t options at their defaults—even though QDAFN often requires more tables and points to get the same quality of results.

$ mlpack_approx_kfn -q queries.csv -r refs.csv -v -k 3 -n n.csv -d d.csv -a qdafn
[INFO ] Loading 'refs.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Building QDAFN model...
[INFO ] Model built.
[INFO ] Loading 'queries.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Searching for 3 furthest neighbors with QDAFN...
[INFO ] Search complete.
[INFO ] Saving CSV data to 'n.csv'.
[INFO ] Saving CSV data to 'd.csv'.
[INFO ]
[INFO ] Execution parameters:
[INFO ]   algorithm: qdafn
[INFO ]   calculate_error: false
[INFO ]   distances_file: d.csv
[INFO ]   exact_distances_file: ""
[INFO ]   help: false
[INFO ]   info: ""
[INFO ]   input_model_file: ""
[INFO ]   k: 3
[INFO ]   neighbors_file: n.csv
[INFO ]   num_projections: 5
[INFO ]   num_tables: 5
[INFO ]   output_model_file: ""
[INFO ]   query_file: queries.csv
[INFO ]   reference_file: refs.csv
[INFO ]   verbose: true
[INFO ]   version: false
[INFO ]
[INFO ] Program timers:
[INFO ]   loading_data: 0.008380s
[INFO ]   qdafn_construct: 0.003399s
[INFO ]   qdafn_search: 0.000886s
[INFO ]   saving_data: 0.002253s
[INFO ]   total_time: 0.015465s

🔗 Printing results quality with exact distances

The mlpack_approx_kfn program can calculate the quality of the results if the --calculate_error (-e) flag is specified. Below we use the program with its default parameters and calculate the error, which is displayed in the output. The error is only calculated for the furthest neighbor, not all k; therefore, in this example we have set -k to 1.

$ mlpack_approx_kfn -q queries.csv -r refs.csv -v -k 1 -e -q -n n.csv
[INFO ] Loading 'refs.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Building DrusillaSelect model...
[INFO ] Model built.
[INFO ] Loading 'queries.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Searching for 1 furthest neighbors with DrusillaSelect...
[INFO ] Search complete.
[INFO ] Calculating exact distances...
[INFO ] 28891 node combinations were scored.
[INFO ] 37735 base cases were calculated.
[INFO ] Calculation complete.
[INFO ] Average error: 1.08417.
[INFO ] Maximum error: 1.28712.
[INFO ] Minimum error: 1.
[INFO ]
[INFO ] Execution parameters:
[INFO ]   algorithm: ds
[INFO ]   calculate_error: true
[INFO ]   distances_file: ""
[INFO ]   exact_distances_file: ""
[INFO ]   help: false
[INFO ]   info: ""
[INFO ]   input_model_file: ""
[INFO ]   k: 3
[INFO ]   neighbors_file: ""
[INFO ]   num_projections: 5
[INFO ]   num_tables: 5
[INFO ]   output_model_file: ""
[INFO ]   query_file: queries.csv
[INFO ]   reference_file: refs.csv
[INFO ]   verbose: true
[INFO ]   version: false
[INFO ]
[INFO ] Program timers:
[INFO ]   computing_neighbors: 0.001476s
[INFO ]   drusilla_select_construct: 0.000309s
[INFO ]   drusilla_select_search: 0.000495s
[INFO ]   loading_data: 0.008462s
[INFO ]   total_time: 0.011670s
[INFO ]   tree_building: 0.000202s

Note that the output includes three lines indicating the error:

[INFO ] Average error: 1.08417.
[INFO ] Maximum error: 1.28712.
[INFO ] Minimum error: 1.

In this case, a minimum error of 1 indicates an exact result, and over the entire query set the algorithm has returned a furthest neighbor candidate with maximum error 1.28712.

🔗 Using cached exact distances for quality results

However, for large datasets, calculating the error may take a long time, because the exact furthest neighbors must be calculated. Therefore, if the exact furthest neighbor distances are already known, they may be passed in with the --exact_distances_file (-x) option in order to avoid the calculation. In the example below, we assume exact.csv contains the exact furthest neighbor distances. We run the qdafn algorithm in this example.

Note that the -e option must be specified for the -x option have any effect.

$ mlpack_approx_kfn -q queries.csv -r refs.csv -k 1 -e -x exact.csv -n n.csv -v -a qdafn
[INFO ] Loading 'refs.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Building QDAFN model...
[INFO ] Model built.
[INFO ] Loading 'queries.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Searching for 1 furthest neighbors with QDAFN...
[INFO ] Search complete.
[INFO ] Loading 'exact.csv' as raw ASCII formatted data.  Size is 1 x 1000.
[INFO ] Average error: 1.06914.
[INFO ] Maximum error: 1.67407.
[INFO ] Minimum error: 1.
[INFO ] Saving CSV data to 'n.csv'.
[INFO ]
[INFO ] Execution parameters:
[INFO ]   algorithm: qdafn
[INFO ]   calculate_error: true
[INFO ]   distances_file: ""
[INFO ]   exact_distances_file: exact.csv
[INFO ]   help: false
[INFO ]   info: ""
[INFO ]   input_model_file: ""
[INFO ]   k: 1
[INFO ]   neighbors_file: n.csv
[INFO ]   num_projections: 5
[INFO ]   num_tables: 5
[INFO ]   output_model_file: ""
[INFO ]   query_file: queries.csv
[INFO ]   reference_file: refs.csv
[INFO ]   verbose: true
[INFO ]   version: false
[INFO ]
[INFO ] Program timers:
[INFO ]   loading_data: 0.010348s
[INFO ]   qdafn_construct: 0.000318s
[INFO ]   qdafn_search: 0.000793s
[INFO ]   saving_data: 0.000259s
[INFO ]   total_time: 0.012254s

🔗 Using tree-based approximation with `mlpack_kfn`

The mlpack_kfn algorithm allows specifying a desired approximation level with the --epsilon (-e) option. The parameter must be greater than or equal to 0 and less than 1. A setting of 0 indicates exact search.

The example below runs dual-tree furthest neighbor search (the default algorithm) with the approximation parameter set to 0.5.

$ mlpack_kfn -q queries.csv -r refs.csv -v -k 3 -e 0.5 -n n.csv -d d.csv
[INFO ] Loading 'refs.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Loaded reference data from 'refs.csv' (3x1000).
[INFO ] Building reference tree...
[INFO ] Tree built.
[INFO ] Loading 'queries.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Loaded query data from 'queries.csv' (3x1000).
[INFO ] Searching for 3 neighbors with dual-tree kd-tree search...
[INFO ] 1611 node combinations were scored.
[INFO ] 13938 base cases were calculated.
[INFO ] 1611 node combinations were scored.
[INFO ] 13938 base cases were calculated.
[INFO ] Search complete.
[INFO ] Saving CSV data to 'n.csv'.
[INFO ] Saving CSV data to 'd.csv'.
[INFO ]
[INFO ] Execution parameters:
[INFO ]   algorithm: dual_tree
[INFO ]   distances_file: d.csv
[INFO ]   epsilon: 0.5
[INFO ]   help: false
[INFO ]   info: ""
[INFO ]   input_model_file: ""
[INFO ]   k: 3
[INFO ]   leaf_size: 20
[INFO ]   naive: false
[INFO ]   neighbors_file: n.csv
[INFO ]   output_model_file: ""
[INFO ]   percentage: 1
[INFO ]   query_file: queries.csv
[INFO ]   random_basis: false
[INFO ]   reference_file: refs.csv
[INFO ]   seed: 0
[INFO ]   single_mode: false
[INFO ]   tree_type: kd
[INFO ]   true_distances_file: ""
[INFO ]   true_neighbors_file: ""
[INFO ]   verbose: true
[INFO ]   version: false
[INFO ]
[INFO ] Program timers:
[INFO ]   computing_neighbors: 0.000442s
[INFO ]   loading_data: 0.008060s
[INFO ]   saving_data: 0.002850s
[INFO ]   total_time: 0.012667s
[INFO ]   tree_building: 0.000251s

Note that the format of the output files d.csv and n.csv are the same as for mlpack_approx_kfn.

🔗 Different algorithms with `mlpack_kfn`

The mlpack_kfn program offers a large number of different algorithms that can be used. The --algorithm (-a) parameter may be used to specify three main different algorithm types: naive (brute-force search), single_tree (single-tree search), dual_tree (dual-tree search, the default), and greedy (“defeatist” greedy search, which goes to one leaf node of the tree then terminates). The example below uses single-tree search to find approximate neighbors with epsilon set to 0.1.

mlpack_kfn -q queries.csv -r refs.csv -v -k 3 -e 0.1 -n n.csv -d d.csv -a single_tree
[INFO ] Loading 'refs.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Loaded reference data from 'refs.csv' (3x1000).
[INFO ] Building reference tree...
[INFO ] Tree built.
[INFO ] Loading 'queries.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Loaded query data from 'queries.csv' (3x1000).
[INFO ] Searching for 3 neighbors with single-tree kd-tree search...
[INFO ] 13240 node combinations were scored.
[INFO ] 15924 base cases were calculated.
[INFO ] Search complete.
[INFO ] Saving CSV data to 'n.csv'.
[INFO ] Saving CSV data to 'd.csv'.
[INFO ]
[INFO ] Execution parameters:
[INFO ]   algorithm: single_tree
[INFO ]   distances_file: d.csv
[INFO ]   epsilon: 0.1
[INFO ]   help: false
[INFO ]   info: ""
[INFO ]   input_model_file: ""
[INFO ]   k: 3
[INFO ]   leaf_size: 20
[INFO ]   naive: false
[INFO ]   neighbors_file: n.csv
[INFO ]   output_model_file: ""
[INFO ]   percentage: 1
[INFO ]   query_file: queries.csv
[INFO ]   random_basis: false
[INFO ]   reference_file: refs.csv
[INFO ]   seed: 0
[INFO ]   single_mode: false
[INFO ]   tree_type: kd
[INFO ]   true_distances_file: ""
[INFO ]   true_neighbors_file: ""
[INFO ]   verbose: true
[INFO ]   version: false
[INFO ]
[INFO ] Program timers:
[INFO ]   computing_neighbors: 0.000850s
[INFO ]   loading_data: 0.007858s
[INFO ]   saving_data: 0.003445s
[INFO ]   total_time: 0.013084s
[INFO ]   tree_building: 0.000250s

🔗 Saving a model for later use

The mlpack_approx_kfn and mlpack_kfn programs both allow models to be saved and loaded for future use. The --output_model_file (-M) option allows specifying where to save a model, and the --input_model_file (-m) option allows a model to be loaded instead of trained. So, if you specify --input_model_file then you do not need to specify --reference_file (-r), --num_projections (-p), or --num_tables (-t).

The example below saves a model with 10 projections and 5 tables. Note that neither --query_file (-q) nor -k are specified; this run only builds the model and saves it to model.bin.

$ mlpack_approx_kfn -r refs.csv -t 5 -p 10 -v -M model.bin
[INFO ] Loading 'refs.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Building DrusillaSelect model...
[INFO ] Model built.
[INFO ]
[INFO ] Execution parameters:
[INFO ]   algorithm: ds
[INFO ]   calculate_error: false
[INFO ]   distances_file: ""
[INFO ]   exact_distances_file: ""
[INFO ]   help: false
[INFO ]   info: ""
[INFO ]   input_model_file: ""
[INFO ]   k: 0
[INFO ]   neighbors_file: ""
[INFO ]   num_projections: 10
[INFO ]   num_tables: 5
[INFO ]   output_model_file: model.bin
[INFO ]   query_file: ""
[INFO ]   reference_file: refs.csv
[INFO ]   verbose: true
[INFO ]   version: false
[INFO ]
[INFO ] Program timers:
[INFO ]   drusilla_select_construct: 0.000321s
[INFO ]   loading_data: 0.004700s
[INFO ]   total_time: 0.007320s

Now, with the model saved, we can run approximate furthest neighbor search on a query set using the saved model:

$ mlpack_approx_kfn -m model.bin -q queries.csv -k 3 -d d.csv -n n.csv -v
[INFO ] Loading 'queries.csv' as CSV data.  Size is 3 x 1000.
[INFO ] Searching for 3 furthest neighbors with DrusillaSelect...
[INFO ] Search complete.
[INFO ] Saving CSV data to 'n.csv'.
[INFO ] Saving CSV data to 'd.csv'.
[INFO ]
[INFO ] Execution parameters:
[INFO ]   algorithm: ds
[INFO ]   calculate_error: false
[INFO ]   distances_file: d.csv
[INFO ]   exact_distances_file: ""
[INFO ]   help: false
[INFO ]   info: ""
[INFO ]   input_model_file: model.bin
[INFO ]   k: 3
[INFO ]   neighbors_file: n.csv
[INFO ]   num_projections: 5
[INFO ]   num_tables: 5
[INFO ]   output_model_file: ""
[INFO ]   query_file: queries.csv
[INFO ]   reference_file: ""
[INFO ]   verbose: true
[INFO ]   version: false
[INFO ]
[INFO ] Program timers:
[INFO ]   drusilla_select_search: 0.000878s
[INFO ]   loading_data: 0.004599s
[INFO ]   saving_data: 0.003006s
[INFO ]   total_time: 0.009234s

These options work in the same way for both the mlpack_approx_kfn and mlpack_kfn programs.

🔗 Final command-line program notes

Both the mlpack_kfn and mlpack_approx_kfn programs contain numerous options not fully documented in these short examples. You can run each program with the --help (-h) option for more information.

🔗 `DrusillaSelect` C++ class

mlpack provides a simple DrusillaSelect C++ class that can be used inside of C++ programs to perform approximate furthest neighbor search. The class has only one template parameter—MatType—which specifies the type of matrix to be use. That means the class can be used with either dense data (of type arma::mat) or sparse data (of type arma::sp_mat).

The following examples show simple usage of this class.

🔗 Approximate furthest neighbors with defaults

The code below builds a DrusillaSelect model with default options on the matrix dataset, then queries for the approximate furthest neighbor of every point in the queries matrix.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::mat dataset;
// The query set.
extern arma::mat queries;

// Construct the model with defaults.
DrusillaSelect<> ds(dataset);

// Query the model, putting output into the following two matrices.
arma::mat distances;
arma::Mat<size_t> neighbors;
ds.Search(queries, 1, neighbors, distances);

At the end of this code, both the distances and neighbors matrices will have number of columns equal to the number of columns in the queries matrix. So, each column of the distances and neighbors matrices are the distances or neighbors of the corresponding column in the queries matrix.

🔗 Custom numbers of tables and projections

The following example constructs a DrusillaSelect model with 10 tables and 5 projections. Once that is done it performs the same task as the previous example.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::mat dataset;
// The query set.
extern arma::mat queries;

// Construct the model with custom parameters.
DrusillaSelect<> ds(dataset, 10, 5);

// Query the model, putting output into the following two matrices.
arma::mat distances;
arma::Mat<size_t> neighbors;
ds.Search(queries, 1, neighbors, distances);

🔗 Accessing the candidate set

The DrusillaSelect algorithm merely scans the reference set and extracts a number of points that will be queried in a brute-force fashion when the Search() method is called. We can access this set with the CandidateSet() method. The code below prints the fifth point of the candidate set.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::mat dataset;

// Construct the model with custom parameters.
DrusillaSelect<> ds(dataset, 10, 5);

// Print the fifth point of the candidate set.
std::cout << ds.CandidateSet().col(4).t();

🔗 Retraining on a new reference set

It is possible to retrain a DrusillaSelect model with new parameters or with a new reference set. This is functionally equivalent to creating a new model. The example code below creates a first DrusillaSelect model using 3 tables and 10 projections, and then retrains this with the same reference set using 10 tables and 3 projections.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::mat dataset;

// Construct the model with initial parameters.
DrusillaSelect<> ds(dataset, 3, 10);

// Now retrain with different parameters.
ds.Train(dataset, 10, 3);

🔗 Running on sparse data

We can set the template parameter for DrusillaSelect to arma::sp_mat in order to perform furthest neighbor search on sparse data. This code below creates a DrusillaSelect model using 4 tables and 6 projections with sparse input data, then searches for 3 approximate furthest neighbors.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::sp_mat dataset;
// The query dataset.
extern arma::sp_mat querySet;

// Construct the model on sparse data.
DrusillaSelect<arma::sp_mat> ds(dataset, 4, 6);

// Search on query data.
arma::Mat<size_t> neighbors;
arma::mat distances;
ds.Search(querySet, 3, neighbors, distances);

🔗 QDAFN C++ class

mlpack also provides a standalone simple QDAFN class for furthest neighbor search. The API for this class is virtually identical to the DrusillaSelect class, and also has one template parameter to specify the type of matrix to be used (dense or sparse or other).

The following subsections demonstrate usage of the QDAFN class in the same way as the previous section’s examples for DrusillaSelect.

🔗 Approximate furthest neighbors with defaults

The code below builds a QDAFN model with default options on the matrix dataset, then queries for the approximate furthest neighbor of every point in the queries matrix.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::mat dataset;
// The query set.
extern arma::mat queries;

// Construct the model with defaults.
QDAFN<> qd(dataset);

// Query the model, putting output into the following two matrices.
arma::mat distances;
arma::Mat<size_t> neighbors;
qd.Search(queries, 1, neighbors, distances);

🔗 Custom numbers of tables and projections

The following example constructs a QDAFN model with 15 tables and 30 projections. Once that is done it performs the same task as the previous example.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::mat dataset;
// The query set.
extern arma::mat queries;

// Construct the model with custom parameters.
QDAFN<> qdafn(dataset, 15, 30);

// Query the model, putting output into the following two matrices.
arma::mat distances;
arma::Mat<size_t> neighbors;
qdafn.Search(queries, 1, neighbors, distances);

🔗 Accessing the candidate set

The QDAFN algorithm scans the reference set, extracting points that have been projected onto random directions. Each random direction corresponds to a single table. The QDAFN class stores these points as a vector of matrices, which can be accessed with the CandidateSet() method. The code below prints the fifth point of the candidate set of the third table.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::mat dataset;

// Construct the model with custom parameters.
QDAFN<> qdafn(dataset, 10, 5);

// Print the fifth point of the candidate set.
std::cout << qdafn.CandidateSet(2).col(4).t();

🔗 Retraining on a new reference set

It is possible to retrain a QDAFN model with new parameters or with a new reference set. This is functionally equivalent to creating a new model. The example code below creates a first QDAFN model using 10 tables and 40 projections, and then retrains this with the same reference set using 15 tables and 25 projections.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::mat dataset;

// Construct the model with initial parameters.
QDAFN<> qdafn(dataset, 3, 10);

// Now retrain with different parameters.
qdafn.Train(dataset, 10, 3);

🔗 Running on sparse data

We can set the template parameter for QDAFN to arma::sp_mat in order to perform furthest neighbor search on sparse data. This code below creates a QDAFN model using 20 tables and 60 projections with sparse input data, then searches for 3 approximate furthest neighbors.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::sp_mat dataset;
// The query dataset.
extern arma::sp_mat querySet;

// Construct the model on sparse data.
QDAFN<arma::sp_mat> qdafn(dataset, 20, 60);

// Search on query data.
arma::Mat<size_t> neighbors;
arma::mat distances;
qdafn.Search(querySet, 3, neighbors, distances);

🔗 KFN C++ class

The extensive NeighborSearch class also provides a way to search for approximate furthest neighbors using a different, tree-based technique. For full documentation on this class, see the NeighborSearch tutorial. The KFN class is a convenient typedef of the NeighborSearch class that can be used to perform the furthest neighbors task with kd-trees.

In the following subsections, the KFN class is used in short code examples.

🔗 Simple furthest neighbors example

The KFN class has construction semantics similar to DrusillaSelect and QDAFN. The example below constructs a KFN object (which will build the tree on the reference set), but note that the third parameter to the constructor allows us to specify our desired level of approximation. In this example we choose epsilon = 0.05. Then, the code searches for 3 approximate furthest neighbors.

#include <mlpack.hpp>

using namespace mlpack;

// The reference dataset.
extern arma::mat dataset;
// The query set.
extern arma::mat querySet;

// Construct the object, performing the default dual-tree search with
// approximation level epsilon = 0.05.
KFN kfn(dataset, DUAL_TREE_MODE, 0.05);

// Search for approximate furthest neighbors.
arma::Mat<size_t> neighbors;
arma::mat distances;
kfn.Search(querySet, 3, neighbors, distances);

🔗 Retraining on a new reference set

Like the QDAFN and DrusillaSelect classes, the KFN class is capable of retraining on a new reference set. The code below demonstrates this.

#include <mlpack.hpp>

using namespace mlpack;

// The original reference set we train on.
extern arma::mat dataset;
// The new reference set we retrain on.
extern arma::mat newDataset;

// Construct the object with approximation level 0.1.
KFN kfn(dataset, DUAL_TREE_MODE, 0.1);

// Retrain on the new reference set.
kfn.Train(newDataset);

🔗 Searching in single-tree mode

The particular mode to be used in search can be specified in the constructor. In this example, we use single-tree search (as opposed to the default of dual-tree search).

#include <mlpack.hpp>

using namespace mlpack;

// The reference set.
extern arma::mat dataset;
// The query set.
extern arma::mat querySet;

// Construct the object with approximation level 0.25 and in single tree search
// mode.
KFN kfn(dataset, SINGLE_TREE_MODE, 0.25);

// Search for 5 approximate furthest neighbors.
arma::Mat<size_t> neighbors;
arma::mat distances;
kfn.Search(querySet, 5, neighbors, distances);

🔗 Searching in brute-force mode

If desired, brute-force search (“naive search”) can be used to find the furthest neighbors; however, the result will not be approximate—it will be exact (since every possibility will be considered). The code below performs exact furthest neighbor search by using the KFN class in brute-force mode.

#include <mlpack.hpp>

using namespace mlpack;

// The reference set.
extern arma::mat dataset;
// The query set.
extern arma::mat querySet;

// Construct the object in brute-force mode.  We can leave the approximation
// parameter to its default (0) since brute-force will provide exact results.
KFN kfn(dataset, NAIVE_MODE);

// Perform the search for 2 furthest neighbors.
arma::Mat<size_t> neighbors;
arma::mat distances;
kfn.Search(querySet, 2, neighbors, distances);

🔗 Further documentation

For further documentation on the approximate furthest neighbor facilities offered by mlpack, see also the NeighborSearch tutorial. Also, each class (QDAFN, DrusillaSelect, NeighborSelect) are well-documented, and more details can be found in the source code documentation.

Approximate furthest neighbor search tutorial

🔗 Which algorithm should be used?

🔗 Command-line mlpack_approx_kfn and mlpack_kfn

🔗 Calculate 5 furthest neighbors with default options

🔗 Specifying algorithm parameters for DrusillaSelect

🔗 Using QDAFN instead of DrusillaSelect

🔗 Printing results quality with exact distances

🔗 Using cached exact distances for quality results

🔗 Using tree-based approximation with mlpack_kfn

🔗 Different algorithms with mlpack_kfn

🔗 Saving a model for later use

🔗 Final command-line program notes

🔗 DrusillaSelect C++ class

🔗 Approximate furthest neighbors with defaults

🔗 Custom numbers of tables and projections

🔗 Accessing the candidate set

🔗 Retraining on a new reference set

🔗 Running on sparse data

🔗 QDAFN C++ class

🔗 Approximate furthest neighbors with defaults

🔗 Custom numbers of tables and projections

🔗 Accessing the candidate set

🔗 Retraining on a new reference set

🔗 Running on sparse data

🔗 KFN C++ class

🔗 Simple furthest neighbors example

🔗 Retraining on a new reference set

🔗 Searching in single-tree mode

🔗 Searching in brute-force mode

🔗 Further documentation

🔗 Command-line `mlpack_approx_kfn` and `mlpack_kfn`

🔗 Specifying algorithm parameters for `DrusillaSelect`

🔗 Using QDAFN instead of `DrusillaSelect`

🔗 Using tree-based approximation with `mlpack_kfn`

🔗 Different algorithms with `mlpack_kfn`

🔗 `DrusillaSelect` C++ class