In [1]:
%matplotlib inline

import matplotlib.pyplot as plt
import seaborn as sns
sns.set(color_codes=True)

from IPython.display import Image

Chapter 9 On-policy Prediction with Approximation

approximate value function: parameterized function $\hat{v}(s, w) \approx v_\pi(s)$

  • applicable to partially observable problems.

9.1 Value-function Approximation

$s \to u$: $s$ is the state updated and $u$ is the update target that $s$'s estimated value is shifted toward.

We use machine learning methods and pass to them the $s \to g$ of each update as a training example. Then we interperet the approximate function they produce as an estimated value function.

not all function approximation methods are equally well suited for use in reinforcement learning:

  • learn efficiently from incrementally acquired data: many traditional methods assume a static training set over which multiple passes are made.
  • are able to handle nonstationary target functions.

9.2 The Prediction Objective (VE)

which states we care most about: a state distribution $\mu(s) \geq 0$, $\sum_s \mu(s) = 1$.

  • Often $\mu(s)$ is chosen to be the fraction of time spent in $s$.

objective function, the Mean Squared Value Error, denoted $\overline{VE}$:

\begin{equation} \overline{VE}(w) \doteq \sum_{s \in \delta} \mu(s) \left [ v_\pi (s) - \hat{v}(s, w) \right ]^2 \end{equation}

where $v_\pi(s)$ is the true value and $\hat{v}(s, w)$ is the approximate value.

Note that best $\overline{VE}$ is no guarantee of our ultimate purpose: to find a better policy.

  • global optimum.
  • local optimum.
  • don't convergence, or diverge.

9.3 Stochastic-gradient and Semi-gradient Methods

SGD: well suited to online reinforcement learning.

\begin{align} w_{t+1} &\doteq w_t - \frac1{2} \alpha \nabla \left [ v_\pi(S_t) - \hat{v}(S_t, w_t) \right ]^2 \\ &= w_t + \alpha \left [ \color{blue}{v_\pi(S_t)} - \hat{v}(S_t, w_t) \right ] \nabla \hat{v}(S_t, w_t) \\ &\approx w_t + \alpha \left [ \color{blue}{U_t} - \hat{v}(S_t, w_t) \right ] \nabla \hat{v}(S_t, w_t) \\ \end{align}

$S_t \to U_t$, is not the true value $v_\pi(S_t)$, but some, possibly random, approximation to it. (前面各种方法累计的value):

  • If $U_t$ is an unbiased estimate, $w_t$ is guaranteed to converge to a local optimum.
  • Otherwise, like boostrappig target or DP target => semi-gradient methods. (might do not converge as robustly as gradient methods)
    • significantly faster learning.
    • enable learning to be continual and online.

state aggregation: states are grouped together, with one estimated value for each group.

9.4 Linear Methods

For every state $s$, there is a real-valued feature vector $x(s) \doteq (x_1(s), x_2(s), \dots, x_d(s))^T$:

\begin{equation} \hat{v}(s, w) \doteq w^T x(s) \doteq \sum_{i=1}^d w_i x_i(s) \end{equation}

9.5 Feature Construction for Linear Methods

Choosing features appropriate to the task is an important way of adding prior domain knowledge to reinforcement learing systems.

  • Polynomials
  • Fourier Basis: low dimension, easy to select, global properities
  • Coarse Coding
  • Tile Coding: convolution kernel?
  • Radial Basis Functions

9.6 Selecting Step-Size Parameters Manually

A good rule of thumb for setting the step-size parameter of linear SGD methods is then $\alpha \doteq (\gamma \mathbf{E}[x^T x])^{-1}$

9.7 Nonlinear Function Approximation: Artificial Neural Networks

+ANN, CNN

9.8 Least-Squares TD

$w_{TD} = A^{-1} b$: data efficient, while expensive computation

9.9 Memory-based Function Approximation

nearest neighbor method

9.10 Kernel-based Function Approximation

RBF function

9.11 Looking Deeper at On-policy Learning: Interest and Emphasis

more interested in some states than others:

  • interest $I_t$: the degree to which we are interested in accurately valuing the state at time $t$.
  • emphaisis $M_t$:

\begin{align} w_{t+n} & \doteq w_{t+n-1} + \alpha M_t \left [ G_{t:t+n} - \hat{v}(S_t, w_{t+n-1} \right ] \nabla \hat{v}(S_t, w_{t+n-1}) \\ M_t & = I_t + \gamma^n M_{t-n}, \qquad 0 \leq t < T \end{align}