Search for: All records

Consider public health officials aiming to spread awareness about a new vaccine in a community interconnected by a social network. How can they distribute information with minimal resources, so as to avoid polarization and ensure community-wide convergence of opinion? To tackle such challenges, we initiate the study of sample complexity of opinion formation in networks. Our framework is built on the recognized opinion formation game, where we regard each agent’s opinion as a data-derived model, unlike previous works that treat opinions as data-independent scalars. The opinion model for every agent is initially learned from its local samples and evolves game-theoretically as all agents communicate with neighbors and revise their models towards an equilibrium. Our focus is on the sample complexity needed to ensure that the opinions converge to an equilibrium such that every agent’s final model has low generalization error. Our paper has two main technical results. First, we present a novel polynomial time optimization framework to quantify the total sample complexity for arbitrary networks, when the underlying learning problem is (generalized) linear regression. Second, we leverage this optimization to study the network gain which measures the improvement of sample complexity when learning over a network compared to that in isolation. Towards this end, we derive network gain bounds for various network classes including cliques, star graphs, and random regular graphs. Additionally, our framework provides a method to study sample distribution within the network, suggesting that it is sufficient to allocate samples inversely to the degree. Empirical results on both synthetic and real-world networks strongly support our theoretical findings. 

Abstract It is natural to generalize the online$$k$$ $k$-Server problem by allowing each request to specify not only a pointp, but also a subsetSof servers that may serve it. To date, only a few special cases of this problem have been studied. The objective of the work presented in this paper has been to more systematically explore this generalization in the case of uniform and star metrics. For uniform metrics, the problem is equivalent to a generalization of Paging in which each request specifies not only a pagep, but also a subsetSof cache slots, and is satisfied by having a copy ofpin some slot inS. We call this problemSlot-Heterogenous Paging. In realistic settings only certain subsets of cache slots or servers would appear in requests. Therefore we parameterize the problem by specifying a family$${\mathcal {S}}\subseteq 2^{[k]}$$ $S\subseteq 2^{\left[k\right]}$of requestable slot sets, and we establish bounds on the competitive ratio as a function of the cache sizekand family$${\mathcal {S}}$$ $S$:If all request sets are allowed ($${\mathcal {S}}=2^{[k]}\setminus \{\emptyset \}$$ $S=2^{\left[k\right]}\backslash \left\{\varnothing \right\}$), the optimal deterministic and randomized competitive ratios are exponentially worse than for standard Paging ($${\mathcal {S}}=\{[k]\}$$ $S=\left\{\right[k\left]\right\}$).As a function of$$|{\mathcal {S}}|$$ $\left|S\right|$andk, the optimal deterministic ratio is polynomial: at most$$O(k^2|{\mathcal {S}}|)$$ $O\left(k^2\right|S\left|\right)$and at least$$\Omega (\sqrt{|{\mathcal {S}}|})$$ $\Omega \left(\sqrt{\left|S\right|}\right)$.For any laminar family$${\mathcal {S}}$$ $S$of heighth, the optimal ratios areO(hk) (deterministic) and$$O(h^2\log k)$$ $O(h^{2}logk)$(randomized).The special case of laminar$${\mathcal {S}}$$ $S$that we callAll-or-One Pagingextends standard Paging by allowing each request to specify a specific slot to put the requested page in. The optimal deterministic ratio forweightedAll-or-One Paging is$$\Theta (k)$$ $\Theta \left(k\right)$. Offline All-or-One Paging is$$\mathbb{N}\mathbb{P}$$ $NP$-hard.Some results for the laminar case are shown via a reduction to the generalization of Paging in which each request specifies a set$$P$$ $P$ofpages, and is satisfied by fetching any page from$$P$$ $P$into the cache. The optimal ratios for the latter problem (with laminar family of heighth) are at mosthk(deterministic) and$$hH_k$$ $h{H}_k$(randomized). 

The classical Erdös-Gallai theorem kicked off the study ofgraph realizability by characterizing degree sequences. We extend this line of research by investigating realizability of directed acyclic graphs (DAGs)given both a local constraint via degree sequences and a global constraint via a sequence of reachability values (number of nodes reachable from a given node). We show that, without degree constraints, DAG reachability realization is solvable in linear time, whereas it is strongly NP-complete given upper bounds on in-degree or out-degree. After defining a suitable notion of bicriteria approximation based on consistency, we give two approximation algorithms achieving O(logn)-reachability consistency and O(logn)-degree consistency; the first, randomized, uses LP (Linear Program) rounding, while the second, deterministic, employs ak-setpacking heuristic. We end with two conjectures that we hope motivate further study of realizability with reachability constraints. 

We initiate the algorithmic study of retracting a graph into a cycle in the graph, which seeks a mapping of the graph vertices to the cycle vertices so as to minimize the maximum stretch of any edge, subject to the constraint that the restriction of the mapping to the cycle is the identity map. This problem has its roots in the rich theory of retraction of topological spaces, and has strong ties to well-studied metric embedding problems such as minimum bandwidth and0-extension. Our first result is anO(min{k,√n})-approximation for retracting any graph on n nodes to a cycle with k nodes. We also show a surprising connection to Sperner’s Lemma that rules out the possibility of improving this result using certain natural convex relaxations of the problem. Nevertheless, if the problem is restricted to planar graphs, we show that we can overcome these integrality gaps by giving an optimal combinatorial algorithm, which is the technical centerpiece of the paper. Building on our planar graph algorithm, we also obtain a constant-factor approximation algorithm for retraction of points in the Euclidean plane to a uniform cycle.