Added implementations.

.gitignore

@@ -0,0 +1,110 @@
+# Data files generated directory
+data/
+
+# Swap files for vim
+*.swp
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# pyenv
+.python-version
+
+# celery beat schedule file
+celerybeat-schedule
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/

README.md

@@ -0,0 +1,67 @@
+# Offline Contextual Bayesian Optimization
+
+## Overview
+
+In Bayesian Optimization (BO), many times there are several systems or "tasks"
+to simultaneously optimize. This repository contains Multi-task Thompson
+Sampling (MTS), a BO algorithm we developed to pick both tasks and actions
+to evaluate. Because some tasks are usually more difficult than others, MTS
+often significantly outperforms standard BO techniques. 
+
+## Getting Set Up
+
+The code is compatible with python 2.7. First, clone this repo and run
+```
+pip install -r requirements
+```
+By default the code leverages the [Dragonfly](https://github.com/dragonfly/dragonfly)
+library. 
+
+## Reproducing Synthetic Experiments
+
+The plots in the paper can be reproduced by running [ocbo.py](src/ocbo.py)
+and [cts_ocbo.py](src/cts_ocbo.py) with the appropriate options file.
+
+```
+cd src
+mkdir data
+python ocbo.py --options <path_to_option_file>
+```
+or if continuous
+```
+python cts_ocbo.py --options <path_to_option_file>
+```
+After the simulation has finished, the plots can be reproduced by
+```
+cd scripts
+python discrete_plotter.py --write_dir ../data --run_id <options_name>
+```
+or
+```
+python cts_plotter.py --write_dir ../data --run_id <options_name>
+```
+For discrete experiments, use the flag `--risk_neutral 1` to show the risk
+neutral performance instead and use `--plot_props 1` flag to show the
+proportion of resources given to different tasks.
+
+With the exception of the experiment in Section 4, the table below shows the
+option file the corresponds to a given experiment.
+| Experiment         | Option File                                           | 
+| -------------      |:-------------:                                        |
+| Figure 1(a,b)      | [set2d.txt](src/options/set2d.txt)                    |
+| Figure 1(c)        | [rand4d.txt](src/options/rand4d.txt)                  |
+| Figure 1(d)        | [rand6d.txt](src/options/rand6d.txt)                  |
+| Figure 1(e)/4(a)   | [jointbran.txt](src/options/jointbran.txt)            |
+| Figure 1(f)/4(b)   | [jointh22.txt](src/options/jointh22.txt)              |
+| Figure 1(g)/4(c)   | [jointh31.txt](src/options/jointh31.txt)              |
+| Figure 1(h)/4(d)   | [jointh42.txt](src/options/jointh42.txt)              |
+| Figure 5(a)        | [contbran.txt](src/options/contbran.txt)              |
+| Figure 5(b)        | [conth22.txt](src/options/conth22.txt)                |
+| Figure 5(c)        | [conth31.txt](src/options/conth31.txt)                |
+| Figure 5(d)        | [conth42.txt](src/options/conth42.txt)                |
+| Figure 5(e)        | [contbran_sethps.txt](src/options/contbran_sethps.txt)|
+| Figure 5(f)        | [conth22_sethps.txt](src/options/conth22_sethps.txt)  |
+| Figure 5(g)        | [conth31_sethps.txt](src/options/conth31_sethps.txt)  |
+| Figure 5(h)        | [conth42_sethps.txt](src/options/conth42_sethps.txt)  |
+
+## Citing Work

requirements.txt

@@ -0,0 +1,18 @@
+backports.functools-lru-cache==1.5
+cycler==0.10.0
+dragonfly-opt==0.1.4
+future==0.18.1
+kiwisolver==1.1.0
+matplotlib==2.2.4
+numpy==1.16.5
+pkg-resources==0.0.0
+pudb==2019.1
+Pygments==2.4.2
+pyparsing==2.4.2
+python-dateutil==2.8.0
+pytz==2019.3
+scipy==1.2.2
+six==1.12.0
+subprocess32==3.5.4
+tqdm==4.36.1
+urwid==2.0.1

src/cstrats/__init__.py

@@ -0,0 +1,17 @@
+"""
+Continuous strategies.
+"""
+from argparse import Namespace
+
+from cstrats.agn_cts import agn_strats, agn_args
+from cstrats.cts_opt import cts_opt_args
+from cstrats.postmax_cts import pm_strats, pm_args
+from cstrats.profile_cts import prof_strats, prof_args
+from cstrats.rand_cts import RandOpt
+
+cstrats = [Namespace(impl=RandOpt, name=RandOpt.get_strat_name())] \
+        + pm_strats \
+        + prof_strats \
+        + agn_strats
+
+copts = cts_opt_args + pm_args + prof_args + agn_args

src/cstrats/agn_cts.py

@@ -0,0 +1,76 @@
+"""
+Randomly select context then optimize.
+"""
+
+from argparse import Namespace
+import numpy as np
+from scipy.stats import norm as normal_distro
+
+from cstrats.cts_opt import ContinuousOpt
+from dragonfly.utils.option_handler import get_option_specs
+from util.misc_util import sample_grid, uniform_draw, knowledge_gradient
+
+agn_args = [\
+        get_option_specs('agn_evals', False, 100,
+            'Number of evaluations for each context to determine max.'),
+]
+
+class AgnosticOpt(ContinuousOpt):
+
+    def _child_set_up(self, function, domain, ctx_dim, options):
+        self.agn_evals = options.agn_evals
+
+    def _determine_next_query(self):
+        # Get the contexts to test out.
+        ctx = self._get_ctx_candidates(1)[0]
+        pt = self._get_best_action(ctx)
+        return pt
+
+    def _get_best_action(self, ctx):
+        """Get the improvement for the context.
+        Args:
+            ctx: ndarray characterizing the context.
+        Returns: Best point.
+        """
+        raise NotImplementedError('Abstract Method')
+
+class AgnEI(AgnosticOpt):
+
+    @staticmethod
+    def get_strat_name():
+        """Get the name of the strategies."""
+        return 'ei'
+
+    def _get_best_action(self, ctx):
+        """Get expected improvement over best posterior mean capped by
+        the best seen reward so far.
+        """
+        act_set = sample_grid([list(ctx)], self.act_domain, self.agn_evals)
+        means, covmat = self.gp.eval(act_set, include_covar=True)
+        best_post = np.max(means)
+        variances = covmat.diagonal().ravel()
+        norm_diff = (means - best_post) / variances
+        eis = norm_diff + normal_distro.cdf(norm_diff) \
+                + normal_distro.pdf(norm_diff)
+        ei_pt = act_set[np.argmax(eis)]
+        return ei_pt
+
+class AgnTS(AgnosticOpt):
+
+    @staticmethod
+    def get_strat_name():
+        """Get the name of the strategies."""
+        return 'ts'
+
+    def _get_best_action(self, ctx):
+        """Get expected improvement over best posterior mean capped by
+        the best seen reward so far.
+        """
+        act_set = sample_grid([list(ctx)], self.act_domain, self.agn_evals)
+        means, covmat = self.gp.eval(act_set, include_covar=True)
+        sample = self.gp.draw_sample(means=means, covar=covmat).ravel()
+        best_pt = act_set[np.argmax(sample)]
+        return best_pt
+
+agn_strats = [Namespace(impl=AgnEI, name=AgnEI.get_strat_name()),
+              Namespace(impl=AgnTS, name=AgnTS.get_strat_name())]