diff options
| -rw-r--r-- | appendix.tex | 62 | ||||
| -rw-r--r-- | approximation.tex | 123 | ||||
| -rw-r--r-- | definitions.tex | 4 |
3 files changed, 109 insertions, 80 deletions
diff --git a/appendix.tex b/appendix.tex index 3685c2f..e964a40 100644 --- a/appendix.tex +++ b/appendix.tex @@ -1,4 +1,5 @@ +\section{Proofs of Statements in Section~\ref{sec:concave}} \subsection{Proof of Lemma~\ref{lemma:relaxation-ratio}}\label{proofofrelaxation-ratio} %\begin{proof} The bound $F_{\mathcal{N}}(\lambda)\leq L_{\mathcal{N}}(\lambda)$ follows by the concavity of the $\log\det$ function and Jensen's inequality. @@ -241,19 +242,20 @@ one fractional component such that \end{equation} Together, \eqref{eq:e1} and \eqref{eq:e2} imply the proposition.\qed -\section{Proof of Proposition~\ref{prop:monotonicity}} +\section{Proof of Proposition~\ref{prop:monotonicity}}\label{proofofpropmonotonicity} -The $\log\det$ function is concave and self-concordant (see -\cite{boyd2004convex}), in this case, the analysis of the barrier method in -in \cite{boyd2004convex} (Section 11.5.5) can be summarized in the following -lemma: +%The $\log\det$ function is concave and self-concordant (see +%\cite{boyd2004convex}), in this case, the analysis of the barrier method in +%in \cite{boyd2004convex} (Section 11.5.5) can be summarized in the following +%lemma: -\begin{lemma}\label{lemma:barrier} -For any $\varepsilon>0$, the barrier method computes an $\varepsilon$-accurate -approximation of $L^*_c$ in time $O(poly(n,d,\log\log\varepsilon^{-1})$. -\end{lemma} +%\begin{lemma}\label{lemma:barrier} +%For any $\varepsilon>0$, the barrier method computes an $\varepsilon$-accurate +%approximation of $L^*_c$ in time $O(poly(n,d,\log\log\varepsilon^{-1})$. +%\end{lemma} +\note{THIBAUT: ARE THE COSTS BELOW DIVIDED BY B? CAN YOU EITHER RENAME THEM OR CARRY THE B AROUND AS APPROPRIATE?} -We show that the optimal value of \eqref{eq:perturbed-primal} is close to the +We proceed by showing that the optimal value of \eqref{eq:perturbed-primal} is close to the optimal value of \eqref{eq:primal} (Lemma~\ref{lemma:proximity}) while being well-behaved with respect to changes of the cost (Lemma~\ref{lemma:monotonicity}). These lemmas together imply @@ -315,7 +317,7 @@ Proposition~\ref{prop:monotonicity}. of the lemma's inequality. \end{proof} -Let us introduce the lagrangian of problem, \eqref{eq:perturbed-primal}: +Let us introduce the Lagrangian of problem, \eqref{eq:perturbed-primal}: \begin{displaymath} \mathcal{L}_{c, \alpha}(\lambda, \mu, \nu, \xi) \defeq L(\lambda) @@ -323,7 +325,7 @@ Let us introduce the lagrangian of problem, \eqref{eq:perturbed-primal}: \end{displaymath} so that: \begin{displaymath} - L^*_c(\alpha) = \min_{\mu, \nu, \xi\geq 0}\max_\lambda \mathcal{L}_{c, \alpha}(\lambda, \mu, \nu, \xi) + L^*_{c,\alpha} = \min_{\mu, \nu, \xi\geq 0}\max_\lambda \mathcal{L}_{c, \alpha}(\lambda, \mu, \nu, \xi) \end{displaymath} Similarly, we define $\mathcal{L}_{c}\defeq\mathcal{L}_{c, 0}$ the lagrangian of \eqref{eq:primal}. @@ -339,19 +341,19 @@ dual feasibility, the KKT conditions give $\forall i\in\{1, \ldots, n\}$: \begin{lemma}\label{lemma:proximity} We have: \begin{displaymath} - L^*_c - \alpha n^2\leq L^*_c(\alpha) \leq L^*_c + L^*_c - \alpha n^2\leq L^*_{c,\alpha} \leq L^*_c \end{displaymath} -In particular, $|L^*_c - L^*_c(\alpha)| \leq \alpha n^2$. +In particular, $|L^*_c - L^*_{c,\alpha}| \leq \alpha n^2$. \end{lemma} \begin{proof} - $\alpha\mapsto L^*_c(\alpha)$ is a decreasing function as it is the + $\alpha\mapsto L^*_{c,\alpha}$ is a decreasing function as it is the maximum value of the $L$ function over a set-decreasing domain, which gives the rightmost inequality. Let $\mu^*, \nu^*, \xi^*$ be dual optimal for $(P_{c, \alpha})$, that is: \begin{displaymath} - L^*_{c}(\alpha) = \max_\lambda \mathcal{L}_{c, \alpha}(\lambda, \mu^*, \nu^*, \xi^*) + L^*_{c,\alpha} = \max_\lambda \mathcal{L}_{c, \alpha}(\lambda, \mu^*, \nu^*, \xi^*) \end{displaymath} Note that $\mathcal{L}_{c, \alpha}(\lambda, \mu^*, \nu^*, \xi^*) @@ -360,11 +362,11 @@ In particular, $|L^*_c - L^*_c(\alpha)| \leq \alpha n^2$. problem \eqref{eq:primal}, $\mathcal{L}_{c}(\lambda, \mu^*, \nu^*, \xi^*) \geq L(\lambda)$. Hence, \begin{displaymath} - L^*_{c}(\alpha) \geq L(\lambda) - \alpha\T{\mathbf{1}}\mu^* + L^*_{c,\alpha} \geq L(\lambda) - \alpha\T{\mathbf{1}}\mu^* \end{displaymath} for any $\lambda$ feasible for \eqref{eq:primal}. In particular, for $\lambda$ primal optimal for $\eqref{eq:primal}$: \begin{equation}\label{eq:local-1} - L^*_{c}(\alpha) \geq L^*_c - \alpha\T{\mathbf{1}}\mu^* + L^*_{c,\alpha} \geq L^*_c - \alpha\T{\mathbf{1}}\mu^* \end{equation} Let us denote by the $M$ the support of $\mu^*$, that is $M\defeq @@ -420,7 +422,7 @@ In particular, $|L^*_c - L^*_c(\alpha)| \leq \alpha n^2$. \begin{lemma}\label{lemma:monotonicity} If $c'$ = $(c_i', c_{-i})$, with $c_i'\leq c_i - \delta$, we have: \begin{displaymath} - L^*_{c'}(\alpha) \geq L^*_c(\alpha) + \frac{\alpha\delta b}{2^n} + L^*_{c',\alpha} \geq L^*_{c,\alpha} + \frac{\alpha\delta b}{2^n} \end{displaymath} \end{lemma} @@ -432,11 +434,11 @@ In particular, $|L^*_c - L^*_c(\alpha)| \leq \alpha n^2$. \end{displaymath} we get similarly to Lemma~\ref{lemma:proximity}: \begin{displaymath} - L^*_{c'}(\alpha) \geq L(\lambda) + \lambda_i\xi^*\delta + L^*_{c',\alpha} \geq L(\lambda) + \lambda_i\xi^*\delta \end{displaymath} for any $\lambda$ feasible for \eqref{eq:perturbed-primal}. In particular, for $\lambda^*$ primal optimal for \eqref{eq:perturbed-primal}: \begin{displaymath} - L^*_{c'}(\alpha) \geq L^*_{c}(\alpha) + \alpha\xi^*\delta + L^*_{c',\alpha} \geq L^*_{c,\alpha} + \alpha\xi^*\delta \end{displaymath} since $\lambda_i^*\geq \alpha$. @@ -447,14 +449,14 @@ In particular, $|L^*_c - L^*_c(\alpha)| \leq \alpha n^2$. with $\lambda^{'*}$ optimal for $(P_{c', \alpha})$. Since $c_i'\leq 1$, using Lemma~\ref{lemma:derivative-bounds}, we get that $\xi^*\geq \frac{b}{2^n}$, which concludes the proof. \end{proof} -\subsection*{End of the proof of Proposition~\ref{prop:monotonicity}} - -Let $\tilde{L}^*_c$ be the approximation computed by +%\subsection*{End of the proof of Proposition~\ref{prop:monotonicity}} +We are now ready to conclude the proof of Proposition~\ref{prop:monotonicity}. +Let $\hat{L}^*_{c,\alpha}$ be the approximation computed by Algorithm~\ref{alg:monotone}. \begin{enumerate} \item using Lemma~\ref{lemma:proximity}: \begin{displaymath} - |\tilde{L}^*_c - L^*_c| \leq |\tilde{L}^*_c - L^*_c(\alpha)| + |L^*_c(\alpha) - L^*_c| + |\hat{L}^*_{c,\alpha} - L^*_c| \leq |\hat{L}^*_{c,\alpha} - L^*_{c,\alpha}| + |L^*_{c,\alpha} - L^*_c| \leq \alpha\delta + \alpha n^2 = \varepsilon \end{displaymath} which proves the $\varepsilon$-accuracy. @@ -462,14 +464,14 @@ which proves the $\varepsilon$-accuracy. \item for the $\delta$-decreasingness, let $c' = (c_i', c_{-i})$ with $c_i'\leq c_i-\delta$, then: \begin{displaymath} - \tilde{L}^*_{c'} \geq L^*_{c'} - \frac{\alpha\delta b}{2^{n+1}} - \geq L^*_c + \frac{\alpha\delta b}{2^{n+1}} - \geq \tilde{L}^*_c + \hat{L}^*_{c',\alpha} \geq L^*_{c',\alpha} - \frac{\alpha\delta b}{2^{n+1}} + \geq L^*_{c,\alpha} + \frac{\alpha\delta b}{2^{n+1}} + \geq \hat{L}^*_{c,\alpha} \end{displaymath} -where the first and inequality come from the accuracy of the approximation, and +where the first and last inequalities follow from the accuracy of the approximation, and the inner inequality follows from Lemma~\ref{lemma:monotonicity}. -\item the accuracy of the approximation $\tilde{L}^*_c$ is: +\item the accuracy of the approximation $\hat{L}^*_{c,\alpha}$ is: \begin{displaymath} A\defeq\frac{\varepsilon\delta b}{2^{n+1}(\delta + n^2)} \end{displaymath} diff --git a/approximation.tex b/approximation.tex index ccd063a..2f73a6b 100644 --- a/approximation.tex +++ b/approximation.tex @@ -50,7 +50,7 @@ expectation of $V$ under the distribution $P_\mathcal{N}^\lambda$: \end{equation} Function $F$ is an extension of $V$ to the domain $[0,1]^n$, as it agrees with $V$ at integer inputs: $F(\id_S) = V(S)$ for all $S\subseteq\mathcal{N}$, where $\id_S$ denotes the indicator vector of $S$. %\citeN{pipage} have shown how to use this extension to obtain approximation guarantees for an interesting class of optimization problems through the \emph{pipage rounding} framework, which has been successfully applied in \citeN{chen, singer-influence}. -Unfortunately, for $V$ the value function given by \eqref{modified} that we study here, the +For $V$ the value function given by \eqref{modified} that we study here, the multi-linear extension \eqref{eq:multi-linear} cannot be computed---let alone maximized---in polynomial time. Hence, we introduce a new extension $L:[0,1]^n\to\reals$: \begin{equation}\label{eq:our-relaxation} @@ -91,9 +91,7 @@ In particular, for $L_c^*\defeq \max_{\lambda\in \dom_c} L(\lambda)$ the optimal \begin{proposition}\label{prop:relaxation} $OPT\leq L^*_c \leq 2 OPT + 2\max_{i\in\mathcal{N}}V(i)$. \end{proposition} -The proof of this proposition can be found in Appendix~\ref{proofofproprelaxation}. Clearly, $L^*_c$ is monotone in $c$: $L^*_c\geq L^*_{c'}$ for any two $c,c'\in \reals_+^n$ s.t.~$c\leq c'$, coordinate-wise. - -%The optimization program \eqref{eq:non-strategic} extends naturally to such +The proof of this proposition can be found in Appendix~\ref{proofofproprelaxation}. As we discuss in the next section, $L^*_c$ can be computed a polynomial time algorithm at arbitrary accuracy; however, the outcome of this computation may not necessarily be monotone: we will address this by converting this poly-time estimator of $L^*_c$ to one that is ``almost'' monotone.%The optimization program \eqref{eq:non-strategic} extends naturally to such %a relaxation. We define: %\begin{equation}\tag{$P_c$}\label{eq:primal} % L^*_c \defeq \max_{\lambda\in[0, 1]^{n}} @@ -109,69 +107,94 @@ The proof of this proposition can be found in Appendix~\ref{proofofproprelaxatio %$OPT$ through rounding. Together, these properties give the following %proposition which is also proved in the Appendix. -\subsection{Solving a Convex Problem Monotonously}\label{sec:monotonicity} +\subsection{Polynomial-Time, Almost-Monotonous Approximation}\label{sec:monotonicity} + The $\log\det$ objective function of \eqref{eq:primal} is known to be concave and \emph{self-concordant} \cite{boyd2004convex}. The maximization of a concave, self-concordant function subject to a set of linear constraints can be performed through the \emph{barrier method} (see, \emph{e.g.}, \cite{boyd2004convex} Section 11.5.5 for general self-concordant optimization as well as \cite{vandenberghe1998determinant} for a detailed treatise of the $\log\det$ case). The performance of the barrier method is summarized in our case by the following lemma: +\begin{lemma}[\citeN{boyd2004convex}]\label{lemma:barrier} +For any $\varepsilon>0$, the barrier method computes an +approximation $\hat{L}^*_c$ that is $\varepsilon$-accurate, \emph{i.e.}, it satisfies $|\hat L^*_c- L^*_c|\leq \varepsilon$, in time $O\left(poly(n,d,\log\log\varepsilon^{-1})\right)$. +\end{lemma} -Note, that the feasible set in Problem~\eqref{eq:primal} increases (for the -inclusion) when the cost decreases. As a consequence, $c\mapsto L^*_c$ is -non-increasing. + Clearly, the optimal value $L^*_c$ of \eqref{eq:primal} is monotone in $c$. Formally, for any two $c,c'\in \reals_+^n$ s.t.~$c\leq c'$ coordinate-wise, $\dom_{c'}\subseteq \dom_c$ and thus $L^*_c\geq L^*_{c'}$. Hence, the map $c\mapsto L^*_c$ is non-increasing. Unfortunately, the same is not true for the output $\hat{L}_c^*$ of the barrier method: there is no guarantee that the $\epsilon$-accurate approximation $\hat{L}^*_c$ will exhibit any kind of monotonicity. -Furthermore, \eqref{eq:primal} being a convex optimization problem, using -standard convex optimization algorithms (Lemma~\ref{lemma:barrier} gives -a formal statement for our specific problem) we can compute -a $\varepsilon$-accurate approximation of its optimal value as defined below. +Nevertheless, it is possible to use the barrier method to construct an approximation of $L^*_{c}$ that is ``almost'' monotone. More specifically, given $\delta>0$, we will say that $f:\reals^n\to\reals$ is +\emph{$\delta$-decreasing} if +\begin{equation}\label{eq:dd} + f(x) \geq f(x+\mu e_i), \qquad + \text{for all}~i\in \mathcal{N},x\in\reals^n, \mu\geq\delta, +\end{equation} +where $e_i$ is the $i$-th canonical basis vector of $\reals^n$. +In other words, $f$ is $\delta$-decreasing if increasing any coordinate by $\delta$ or more at input $x$ ensures that the input will be at most $f(x)$. -\begin{definition} -$a$ is an $\varepsilon$-accurate approximation of $b$ iff $|a-b|\leq \varepsilon$. -\end{definition} +Our next technical lemma establishes that, using the barrier method, it is possible to construct an algorithm that computes $L^*_c$ at arbitrary accuracy in polynomial time \emph{and} is $\delta$-monotone. We achieve this by restricting the optimization over a subset of $\dom_c$ at which the concave relaxation $L$ is ``sufficiently'' concave. Formally, for $\alpha\geq 0$ let $$\textstyle\dom_{c,\alpha} \defeq \{\lambda \in [\alpha,1]^n: \sum_{i\in \mathcal{N}}c_i\lambda_i \leq B\}\subseteq \dom_c . $$ +Note that $\dom_c=\dom_{c,0}.$ Consider the following perturbation of the concave relaxation \eqref{eq:primal}: +\begin{align}\tag{$P_{c,\alpha}$}\label{eq:perturbed-primal} +\begin{split} \text{Maximize:} &\qquad L(\lambda)\\ +\text{subject to:} & \qquad\lambda \in \dom_{c,\alpha} +\end{split} +\end{align} -Note however that an $\varepsilon$-accurate approximation of a non-increasing -function is not in general non-increasing itself. The goal of this section is -to approximate $L^*_c$ while preserving monotonicity. The estimator we -construct has a weaker form of monotonicity that we call -\emph{$\delta$-monotonicity}. +%Note, that the feasible set in Problem~\eqref{eq:primal} increases (for the +%inclusion) when the cost decreases. +%non-increasing. -\begin{definition} -Let $f$ be a function from $\reals^n$ to $\reals$, we say that $f$ is -\emph{$\delta$-increasing along the $i$-th coordinate} iff: -\begin{equation}\label{eq:dd} - \forall x\in\reals^n,\; - \forall \mu\geq\delta,\; - f(x+\mu e_i)\geq f(x) -\end{equation} -where $e_i$ is the $i$-th canonical basis vector of $\reals^n$. By extension, -$f$ is $\delta$-increasing iff it is $\delta$-increasing along all coordinates. +%Furthermore, \eqref{eq:primal} being a convex optimization problem, using +%standard convex optimization algorithms (Lemma~\ref{lemma:barrier} gives +%a formal statement for our specific problem) we can compute +%a $\varepsilon$-accurate approximation of its optimal value as defined below. -We define \emph{$\delta$-decreasing} functions by reversing the inequality in -\eqref{eq:dd}. -\end{definition} +%\begin{definition} +%$a$ is an $\varepsilon$-accurate approximation of $b$ iff $|a-b|\leq \varepsilon$. +%\end{definition} -We consider a perturbation of \eqref{eq:primal} by introducing: -\begin{equation}\tag{$P_{c, \alpha}$}\label{eq:perturbed-primal} - L^*_c(\alpha) \defeq \max_{\lambda\in[\alpha, 1]^{n}} - \left\{L(\lambda) \Big| \sum_{i=1}^{n} \lambda_i c_i - \leq B\right\} -\end{equation} -Note that we have $L^*_c = L^*_c(0)$. We will also assume that -$\alpha<\frac{1}{nB}$ so that \eqref{eq:perturbed-primal} has at least one -feasible point: $(\frac{1}{nB},\ldots,\frac{1}{nB})$. By computing -an approximation of $L^*_c(\alpha)$ as in Algorithm~\ref{alg:monotone}, we -obtain a $\delta$-decreasing approximation of $L^*_c$. +%Note however that an $\varepsilon$-accurate approximation of a non-increasing +%function is not in general non-increasing itself. The goal of this section is +%to approximate $L^*_c$ while preserving monotonicity. The estimator we +%construct has a weaker form of monotonicity that we call +%\emph{$\delta$-monotonicity}. + +%\begin{definition} +%Let $f$ be a function from $\reals^n$ to $\reals$, we say that $f$ is +%\emph{$\delta$-increasing along the $i$-th coordinate} iff: +%\begin{equation}\label{eq:dd} +% \forall x\in\reals^n,\; +% \forall \mu\geq\delta,\; +% f(x+\mu e_i)\geq f(x) +%\end{equation} +%where $e_i$ is the $i$-th canonical basis vector of $\reals^n$. By extension, +%$f$ is $\delta$-increasing iff it is $\delta$-increasing along all coordinates. + +%We define \emph{$\delta$-decreasing} functions by reversing the inequality in +%\eqref{eq:dd}. +%\end{definition} + +%We consider a perturbation of \eqref{eq:primal} by introducing: +%\begin{equation}\tag{$P_{c, \alpha}$}\label{eq:perturbed-primal} +% L^*_{c,\alpha} \defeq \max_{\lambda\in[\alpha, 1]^{n}} +% \left\{L(\lambda) \Big| \sum_{i=1}^{n} \lambda_i c_i +% \leq B\right\} +%\end{equation} +%Note that we have $L^*_c = L^*_c(0)$. We will also assume that +%$\alpha<\frac{1}{nB}$ so that \eqref{eq:perturbed-primal} has at least one +%feasible point: $(\frac{1}{nB},\ldots,\frac{1}{nB})$. By computing +%an approximation of $L^*_{c,\alpha}$ as in Algorithm~\ref{alg:monotone}, we +%obtain a $\delta$-decreasing approximation of $L^*_c$. \begin{algorithm}[t] \caption{}\label{alg:monotone} \begin{algorithmic}[1] - \State $\alpha \gets \varepsilon B^{-1}(\delta+n^2)^{-1} $ - - \State Compute a $\frac{1}{2^{n+1}}\alpha\delta b$-accurate approximation of - $L^*_c(\alpha)$ + \Require{ $B\in \reals_+$,$c\in\reals^n_+$, $\delta\in (0,1]$, $\epsilon\in (0,1]$ } + \State $\alpha \gets \min\left(\varepsilon B^{-1}(\delta+n^2)^{-1},\frac{1}{nB}\right) $ + \State Use the barrier method to solve \eqref{eq:perturbed-primal} with accuracy $\varepsilon'=\frac{1}{2^{n+1}}\alpha\delta b$; denote the output by $\hat{L}^*_{c,\alpha}$ + \State \textbf{return} $\hat{L}^*_{c,\alpha}$ \end{algorithmic} \end{algorithm} +Our construction of a $\delta$-decreasing, $\varepsilon$-accurate approximator of $\hat{L}_c^*$ proceeds as follows: first, it computes an appropriately selected lower bound $\alpha$; using this bound, it proceeds by solving the perturbed problem \eqref{eq:perturbed-primal} using the barrier method, also at an appropriately selected accuracy $\varepsilon'$. The corresponding output is returned as an approximation of $L^*_c$. A summary of algorithm and the specific choices of $\alpha$ and $\varepsilon'$ are given in Algorithm~\ref{alg:monotone}. The following proposition, which is proved in Appendix~\ref{proofofpropmonotonicity}, establishes that this algorithm has the desirable properties: \begin{proposition}\label{prop:monotonicity} For any $\delta\in(0,1]$ and any $\varepsilon\in(0,1]$, Algorithm~\ref{alg:monotone} computes a $\delta$-decreasing, $\varepsilon$-accurate approximation of $L^*_c$. The running time of the - algorithm is $O\big(poly(n, d, \log\log\frac{1}{b\varepsilon\delta})\big)$ + algorithm is $O\big(poly(n, d, \log\log\frac{1}{b\varepsilon\delta})\big)$. \note{THIBAUT, IS THERE A BUDGET $B$ IN THE DENOM OF THE LOGLOG TERM?} \end{proposition} - +We note that the the execution of the barrier method on the restricted set $\dom_{c,\alpha}$ is necessary. The algorithm's output when executed over the entire domain may not necessarily be $\delta$-decreasing, even when the approximation accuracy is small. This is because when the optimal $\lambda\in \dom_c$ lies at the boundary, certain costs can be saturated: increasing them has no effect on the objective. Forcing the optimization to happen ``off'' the boundary ensures that this does not occur, while taking $\alpha$ to be small ensures that this perturbation does not cost much in terms of approximation accuracy. diff --git a/definitions.tex b/definitions.tex index 555c02e..a719fce 100644 --- a/definitions.tex +++ b/definitions.tex @@ -35,3 +35,7 @@ \newcommand{\junk}[1]{} \newcommand{\edp}{{\tt EDP}} \newcommand{\dom}{\mathcal{D}} +\newcommand{\note}[1]{\textcolor{red}{#1}} + +\algrenewcommand\algorithmicrequire{\textbf{Input:}} +\algrenewcommand\algorithmicensure{\textbf{Output:}} |