VROOM algorithm implementation #33 #6

README.md

@@ -134,6 +134,7 @@ pip install .
 | [PCT](https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/PCT.py)             | [General Parallel Optimization Without A Metric](https://proceedings.mlr.press/v98/xuedong19a.html)                                                                                      | 2019 |
 | [SequOOL](https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/SequOOL.py)     | [A Simple Parameter-free And Adaptive Approach to Optimization Under A Minimal Local Smoothness Assumption](https://arxiv.org/pdf/1810.00997.pdf)                                        | 2019 |
 | [StroquOOL](https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/StroquOOL.py) | [A Simple Parameter-free And Adaptive Approach to Optimization Under A Minimal Local Smoothness Assumption](https://arxiv.org/pdf/1810.00997.pdf)                                        | 2019 |
+| [VROOM](https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/VROOM.py)         | [Derivative-Free & Order-Robust Optimisation](https://arxiv.org/pdf/1910.04034.pdf)                                                                                                      | 2020 |
 | [VHCT](https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/VHCT.py)           | [Optimum-statistical Collaboration Towards General and Efficient Black-box Optimization](https://openreview.net/forum?id=ClIcmwdlxn)                                                     | 2023 |
 | [VPCT](https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/VPCT.py)           | N.A. ([GPO](https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/GPO.py) + [VHCT](https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/VHCT.py))                               | N.A. |
 

docs/source/api/algorithms.rst

@@ -181,6 +181,24 @@ StroquOOL Algorithm
     :special-members: __init__
 
 
+..................................
+
+VROOM Algorithm
+----------------------------
+.. autoclass:: PyXAB.algos.VROOM.VROOM_node
+    :members:
+    :undoc-members:
+    :show-inheritance:
+    :inherited-members:
+    :special-members: __init__
+
+.. autoclass:: PyXAB.algos.VROOM.VROOM
+    :members:
+    :undoc-members:
+    :show-inheritance:
+    :inherited-members:
+    :special-members: __init__
+
 
 ..................................
 

docs/source/features/algorithms/VROOM/VROOM.rst

@@ -0,0 +1,52 @@
+VROOM Algorithm
+===============
+
+Introduction
+------------
+`paper <https://proceedings.mlr.press/v108/ammar20a/ammar20a.pdf>`_, `code <https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/VROOM.py>`_
+
+In this paper, we formalise order-robust optimisation as an instance of online learning minimising simple regret,
+and propose VROOM, a zeroth order optimisation algorithm capable of achieving vanishing regret in non-stationary
+environments, while recovering favorable rates under stochastic reward-generating processes. Our results are the first
+to target simple regret definitions in adversarial scenarios unveiling a challenge that has been rarely considered
+in prior work.
+
+Algorithm Parameters
+--------------------
+    * n: int
+        The total number of rounds (budget)
+    * h_max: int
+        The maximum depth of the partition
+    * b: float
+        The parameter that measures the variation of the function
+    * f_max: float
+        An upper bound of the objective function
+    * domain: list(list)
+        The domain of the objective to be optimized
+    * partition:
+        The partition choice of the algorithm
+
+Usage Example
+-------------
+
+.. note::
+
+    Make sure to use `get_last_point()` to get the final output
+
+
+.. code-block:: python3
+
+    from PyXAB.synthetic_obj.Garland import Garland
+    from PyXAB.algos.VROOM import VROOM
+
+    domain = [[0, 1]]               # Parameter is 1-D and between 0 and 1
+    target = Garland()
+    rounds = 1000
+    algo = VROOM(n=rounds, b=1, domain=domain)
+
+    for t in range(rounds):
+        point = algo.pull(t)
+        reward = target(point)
+        algo.receive_reward(t, reward)
+
+    algo.get_last_point()

docs/source/features/feat_algorithms.rst

@@ -84,6 +84,11 @@ features in the algorithm design.
      - |check_|
      - |cross_|
      - |cross_|
+   * - `VROOM <https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/VROOM.py>`_
+     - `VROOM paper <https://arxiv.org/pdf/1910.04034.pdf>`_
+     - |check_|
+     - |cross_|
+     - |cross_|
    * - `VHCT <https://github.com/WilliamLwj/PyXAB/blob/main/PyXAB/algos/VHCT.py>`_
      - `VHCT paper <https://openreview.net/forum?id=ClIcmwdlxn>`_
      - |check_|
@@ -123,6 +128,7 @@ features in the algorithm design.
     algorithms/PCT/PCT
     algorithms/SequOOL/SequOOL
     algorithms/StroquOOL/StroquOOL
+    algorithms/VROOM/VROOM
     algorithms/VHCT/VHCT
     algorithms/VPCT/VPCT