Graph.anchored_difference_bounded_variables(count, lower_bound, upper_bound, difference_bound, initial_values=None, *, name=None)

Create a sequence of optimizable variables with an anchored difference bound.

Use this function to create a sequence of optimization variables that have a difference bound (each variable is constrained to be within a given distance of the adjacent variables) and are anchored to zero at the start and end (the initial and final variables are within a given distance of zero).

  • count (int) – The number \(N\) of individual real-valued variables to create.

  • lower_bound (float) – The lower bound \(v_\mathrm{min}\) on the variables. The same lower bound applies to all count individual variables.

  • upper_bound (float) – The upper bound \(v_\mathrm{max}\) on the variables. The same upper bound applies to all count individual variables.

  • difference_bound (float) – The difference bound \(\delta\) to enforce between adjacent variables.

  • initial_values (np.ndarray or List[np.ndarray] or None, optional) – Initial values for optimization variable. You can either provide a single initial value, or a list of them. Note that all optimization variables in a graph with non-default initial values must have the same length. That is, you must set them either as a single array or a list of arrays of the same length. Defaults to None, meaning the optimizer initializes the variables with random values.

  • name (str or None, optional) – The name of the node.


The sequence \(\{v_n\}\) of \(N\) anchored difference-bounded optimization variables, satisfying \(v_\mathrm{min}\leq v_n\leq v_\mathrm{max}\), \(|v_{n-1}-v_n|\leq\delta\) for \(2\leq n\leq N\), \(|v_1|\leq\delta\), and \(|v_N|\leq\delta\).

Return type:


See also


Create 1D Tensor of optimization variables.


Function to find the minimum of generic deterministic functions.


Function to find the minimum of generic stochastic functions.


Create optimizable PWC signal with anchored difference bound.

>>> values = graph.anchored_difference_bounded_variables(
...     count=10, lower_bound=-1, upper_bound=1, difference_bound=0.1
... )
>>> graph.pwc_signal(values=values, duration=1)
<Pwc: name="pwc_signal_#1", operation_name="pwc_signal", value_shape=(), batch_shape=()>

See the “Band-limited pulses with bounded slew rates” example in the How to add smoothing and band-limits to optimized controls user guide.