Scippy

SCIP

Solving Constraint Integer Programs

pub_bandit_ucb.h
Go to the documentation of this file.
1 /* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */
2 /* */
3 /* This file is part of the program and library */
4 /* SCIP --- Solving Constraint Integer Programs */
5 /* */
6 /* Copyright 2002-2022 Zuse Institute Berlin */
7 /* */
8 /* Licensed under the Apache License, Version 2.0 (the "License"); */
9 /* you may not use this file except in compliance with the License. */
10 /* You may obtain a copy of the License at */
11 /* */
12 /* http://www.apache.org/licenses/LICENSE-2.0 */
13 /* */
14 /* Unless required by applicable law or agreed to in writing, software */
15 /* distributed under the License is distributed on an "AS IS" BASIS, */
16 /* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. */
17 /* See the License for the specific language governing permissions and */
18 /* limitations under the License. */
19 /* */
20 /* You should have received a copy of the Apache-2.0 license */
21 /* along with SCIP; see the file LICENSE. If not visit scipopt.org. */
22 /* */
23 /* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */
24 
25 /**@file pub_bandit_ucb.h
26  * @ingroup PublicBanditMethods
27  * @brief public methods for UCB bandit selection
28  * @author Gregor Hendel
29  */
30 
31 /*---+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8----+----9----+----0----+----1----+----2*/
32 
33 #ifndef SRC_SCIP_PUB_BANDIT_UCB_H_
34 #define SRC_SCIP_PUB_BANDIT_UCB_H_
35 
36 #include "scip/def.h"
37 #include "scip/type_bandit.h"
38 #include "scip/type_retcode.h"
39 #include "scip/type_scip.h"
40 
41 #ifdef __cplusplus
42 extern "C" {
43 #endif
44 
45 
46 /**@addtogroup PublicBanditMethods
47  *
48  * ## Upper Confidence Bounds (UCB)
49  *
50  * UCB (Upper confidence bounds) is a deterministic
51  * selection algorithm for the multi-armed bandit problem.
52  * In every iteration, UCB selects the action that maximizes
53  * a tradeoff between its performance in the past
54  * and a variance term.
55  * The influence of the variance (confidence width) can be
56  * controlled by the parameter \f$ \alpha \f$.
57  *
58  * @{
59  */
60 
61 
62 /** create and reset UCB bandit algorithm */
63 SCIP_EXPORT
65  SCIP* scip, /**< SCIP data structure */
66  SCIP_BANDIT** ucb, /**< pointer to store bandit algorithm */
67  SCIP_Real* priorities, /**< nonnegative priorities for each action, or NULL if not needed */
68  SCIP_Real alpha, /**< parameter to increase confidence width */
69  int nactions, /**< the positive number of actions for this bandit algorithm */
70  unsigned int initseed /**< initial random number seed */
71  );
72 
73 /** returns the upper confidence bound of a selected action */
74 SCIP_EXPORT
76  SCIP_BANDIT* ucb, /**< UCB bandit algorithm */
77  int action /**< index of the queried action */
78  );
79 
80 /** return start permutation of the UCB bandit algorithm */
81 SCIP_EXPORT
83  SCIP_BANDIT* ucb /**< UCB bandit algorithm */
84  );
85 
86 /** @}*/
87 
88 
89 #ifdef __cplusplus
90 }
91 #endif
92 
93 #endif
SCIP_RETCODE SCIPcreateBanditUcb(SCIP *scip, SCIP_BANDIT **ucb, SCIP_Real *priorities, SCIP_Real alpha, int nactions, unsigned int initseed)
Definition: bandit_ucb.c:339
enum SCIP_Retcode SCIP_RETCODE
Definition: type_retcode.h:63
type definitions for return codes for SCIP methods
type definitions for SCIP&#39;s main datastructure
type definitions for bandit selection algorithms
int * SCIPgetStartPermutationUcb(SCIP_BANDIT *ucb)
Definition: bandit_ucb.c:293
#define SCIP_Real
Definition: def.h:186
common defines and data types used in all packages of SCIP
SCIP_Real SCIPgetConfidenceBoundUcb(SCIP_BANDIT *ucb, int action)
Definition: bandit_ucb.c:264