arXiv subjects

Binghui Peng

Publications and source records attributed to Binghui Peng.

32 records · Page 2Linked to original sources

arXiv2022

Continual learning: a feature extraction formalization, an efficient algorithm, and fundamental obstructions

Continual learning is an emerging paradigm in machine learning, wherein a model is exposed in an online fashion to data from multiple different distributions (i.e. environments), and is expected to adapt to the distribution change. Precisely, the goal is to perform well in the new environment, while simultaneously retaining the performance on the previous environments (i.e. avoid "catastrophic forgetting") -- without increasing the size of the model. While this setup has enjoyed a lot of attention in the applied community, there hasn't be theoretical work that even formalizes the desired guarantees. In this paper, we propose a framework for continual learning through the framework of feature extraction -- namely, one in which features, as well as a classifier, are being trained with each environment. When the features are linear, we design an efficient gradient-based algorithm $\mathsf{DPGD}$, that is guaranteed to perform well on the current environment, as well as avoid catastrophic forgetting. In the general case, when the features are non-linear, we show such an algorithm cannot exist, whether efficient or not.

cs.LG↗

arXiv2022

Shuffle Private Stochastic Convex Optimization

In shuffle privacy, each user sends a collection of randomized messages to a trusted shuffler, the shuffler randomly permutes these messages, and the resulting shuffled collection of messages must satisfy differential privacy. Prior work in this model has largely focused on protocols that use a single round of communication to compute algorithmic primitives like means, histograms, and counts. We present interactive shuffle protocols for stochastic convex optimization. Our protocols rely on a new noninteractive protocol for summing vectors of bounded $\ell_2$ norm. By combining this sum subroutine with mini-batch stochastic gradient descent, accelerated gradient descent, and Nesterov's smoothing method, we obtain loss guarantees for a variety of convex loss functions that significantly improve on those of the local model and sometimes match those of the central model.

cs.LG↗

arXiv2021

Dynamic Influence Maximization

We initiate a systematic study on $\mathit{dynamic}$ $\mathit{influence}$ $\mathit{maximization}$ (DIM). In the DIM problem, one maintains a seed set $S$ of at most $k$ nodes in a dynamically involving social network, with the goal of maximizing the expected influence spread while minimizing the amortized updating cost. We consider two evolution models. In the $\mathit{incremental}$ model, the social network gets enlarged over time and one only introduces new users and establishes new social links, we design an algorithm that achieves $(1-1/e-ε)$-approximation to the optimal solution and has $k \cdot\mathsf{poly}(\log n, ε^{-1})$ amortized running time, which matches the state-of-art offline algorithm with only poly-logarithmic overhead. In the $\mathit{fully}$ $\mathit{dynamic}$ model, users join in and leave, influence propagation gets strengthened or weakened in real time, we prove that under the Strong Exponential Time Hypothesis (SETH), no algorithm can achieve $2^{-(\log n)^{1-o(1)}}$-approximation unless the amortized running time is $n^{1-o(1)}$. On the technical side, we exploit novel adaptive sampling approaches that reduce DIM to the dynamic MAX-k coverage problem, and design an efficient $(1-1/e-ε)$-approximation algorithm for it. Our lower bound leverages the recent developed distributed PCP framework.

cs.DS↗

arXiv2021

Public Goods Games in Directed Networks

Public goods games in undirected networks are generally known to have pure Nash equilibria, which are easy to find. In contrast, we prove that, in directed networks, a broad range of public goods games have intractable equilibrium problems: The existence of pure Nash equilibria is NP-hard to decide, and mixed Nash equilibria are PPAD-hard to find. We define general utility public goods games, and prove a complexity dichotomy result for finding pure equilibria, and a PPAD-completeness proof for mixed Nash equilibria. Even in the divisible goods variant of the problem, where existence is easy to prove, finding the equilibrium is PPAD-complete. Finally, when the treewidth of the directed network is appropriately bounded, we prove that polynomial-time algorithms are possible.

cs.GT↗

arXiv2021

Computational Hardness of the Hylland-Zeckhauser Scheme

We study the complexity of the classic Hylland-Zeckhauser scheme [HZ'79] for one-sided matching markets. We show that the problem of finding an $ε$-approximate equilibrium in the HZ scheme is PPAD-hard, and this holds even when $ε$ is polynomially small and when each agent has no more than four distinct utility values. Our hardness result, when combined with the PPAD membership result of [VY'21], resolves the approximation complexity of the HZ scheme. We also show that the problem of approximating the optimal social welfare (the weight of the matching) achievable by HZ equilibria within a certain constant factor is NP-hard.

cs.GT↗

arXiv2020

Training (Overparametrized) Neural Networks in Near-Linear Time

The slow convergence rate and pathological curvature issues of first-order gradient methods for training deep neural networks, initiated an ongoing effort for developing faster $\mathit{second}$-$\mathit{order}$ optimization algorithms beyond SGD, without compromising the generalization error. Despite their remarkable convergence rate ($\mathit{independent}$ of the training batch size $n$), second-order algorithms incur a daunting slowdown in the $\mathit{cost}$ $\mathit{per}$ $\mathit{iteration}$ (inverting the Hessian matrix of the loss function), which renders them impractical. Very recently, this computational overhead was mitigated by the works of [ZMG19,CGH+19}, yielding an $O(mn^2)$-time second-order algorithm for training two-layer overparametrized neural networks of polynomial width $m$. We show how to speed up the algorithm of [CGH+19], achieving an $\tilde{O}(mn)$-time backpropagation algorithm for training (mildly overparametrized) ReLU networks, which is near-linear in the dimension ($mn$) of the full gradient (Jacobian) matrix. The centerpiece of our algorithm is to reformulate the Gauss-Newton iteration as an $\ell_2$-regression problem, and then use a Fast-JL type dimension reduction to $\mathit{precondition}$ the underlying Gram matrix in time independent of $M$, allowing to find a sufficiently good approximate solution via $\mathit{first}$-$\mathit{order}$ conjugate gradient. Our result provides a proof-of-concept that advanced machinery from randomized linear algebra -- which led to recent breakthroughs in $\mathit{convex}$ $\mathit{optimization}$ (ERM, LPs, Regression) -- can be carried over to the realm of deep learning as well.

cs.LG↗

arXiv2020

MixCon: Adjusting the Separability of Data Representations for Harder Data Recovery

To address the issue that deep neural networks (DNNs) are vulnerable to model inversion attacks, we design an objective function, which adjusts the separability of the hidden data representations, as a way to control the trade-off between data utility and vulnerability to inversion attacks. Our method is motivated by the theoretical insights of data separability in neural networking training and results on the hardness of model inversion. Empirically, by adjusting the separability of data representation, we show that there exist sweet-spots for data separability such that it is difficult to recover data during inference while maintaining data utility.

cs.LG↗

arXiv2020

Hedging in games: Faster convergence of external and swap regrets

We consider the setting where players run the Hedge algorithm or its optimistic variant to play an $n$-action game repeatedly for $T$ rounds. 1) For two-player games, we show that the regret of optimistic Hedge decays at $\tilde{O}( 1/T ^{5/6} )$, improving the previous bound $O(1/T^{3/4})$ by Syrgkanis, Agarwal, Luo and Schapire (NIPS'15) 2) In contrast, we show that the convergence rate of vanilla Hedge is no better than $\tildeΩ(1/ \sqrt{T})$, addressing an open question posted in Syrgkanis, Agarwal, Luo and Schapire (NIPS'15). For general m-player games, we show that the swap regret of each player decays at rate $\tilde{O}(m^{1/2} (n/T)^{3/4})$ when they combine optimistic Hedge with the classical external-to-internal reduction of Blum and Mansour (JMLR'07). The algorithm can also be modified to achieve the same rate against itself and a rate of $\tilde{O}(\sqrt{n/T})$ against adversaries. Via standard connections, our upper bounds also imply faster convergence to coarse correlated equilibria in two-player games and to correlated equilibria in multiplayer games.

cs.GT↗

arXiv2019

On Adaptivity Gaps of Influence Maximization under the Independent Cascade Model with Full Adoption Feedback

In this paper, we study the adaptivity gap of the influence maximization problem under independent cascade model when full-adoption feedback is available. Our main results are to derive upper bounds on several families of well-studied influence graphs, including in-arborescences, out-arborescences and bipartite graphs. Especially, we prove that the adaptivity gap for the in-arborescence is between $[\frac{e}{e-1}, \frac{2e}{e - 1}]$ and for the out-arborescence, the gap is between $[\frac{e}{e-1}, 2]$. These are the first constant upper bounds in the full-adoption feedback model. We provide several novel ideas to tackle with correlated feedback appearing in the adaptive stochastic optimization, which we believe to be of independent interests.

cs.SI↗

arXiv2019

Adaptive Influence Maximization with Myopic Feedback

We study the adaptive influence maximization problem with myopic feedback under the independent cascade model: one sequentially selects k nodes as seeds one by one from a social network, and each selected seed returns the immediate neighbors it activates as the feedback available for later selections, and the goal is to maximize the expected number of total activated nodes, referred as the influence spread. We show that the adaptivity gap, the ratio between the optimal adaptive influence spread and the optimal non-adaptive influence spread, is at most 4 and at least e/(e-1), and the approximation ratios with respect to the optimal adaptive influence spread of both the non-adaptive greedy and adaptive greedy algorithms are at least \frac{1}{4}(1 - \frac{1}{e}) and at most \frac{e^2 + 1}{(e + 1)^2} < 1 - \frac{1}{e}. Moreover, the approximation ratio of the non-adaptive greedy algorithm is no worse than that of the adaptive greedy algorithm, when considering all graphs. Our result confirms a long-standing open conjecture of Golovin and Krause (2011) on the constant approximation ratio of adaptive greedy with myopic feedback, and it also suggests that adaptive greedy may not bring much benefit under myopic feedback.

cs.SI↗

arXiv2019

Tight Bounds for Online Edge Coloring

Vizing's celebrated theorem asserts that any graph of maximum degree $Δ$ admits an edge coloring using at most $Δ+1$ colors. In contrast, Bar-Noy, Naor and Motwani showed over a quarter century that the trivial greedy algorithm, which uses $2Δ-1$ colors, is optimal among online algorithms. Their lower bound has a caveat, however: it only applies to low-degree graphs, with $Δ=O(\log n)$, and they conjectured the existence of online algorithms using $Δ(1+o(1))$ colors for $Δ=ω(\log n)$. Progress towards resolving this conjecture was only made under stochastic arrivals (Aggarwal et al., FOCS'03 and Bahmani et al., SODA'10). We resolve the above conjecture for \emph{adversarial} vertex arrivals in bipartite graphs, for which we present a $(1+o(1))Δ$-edge-coloring algorithm for $Δ=ω(\log n)$ known a priori. Surprisingly, if $Δ$ is not known ahead of time, we show that no $\big(\frac{e}{e-1} - Ω(1) \big) Δ$-edge-coloring algorithm exists. We then provide an optimal, $\big(\frac{e}{e-1}+o(1)\big)Δ$-edge-coloring algorithm for unknown $Δ=ω(\log n)$. Key to our results, and of possible independent interest, is a novel fractional relaxation for edge coloring, for which we present optimal fractional online algorithms and a near-lossless online rounding scheme, yielding our optimal randomized algorithms.

cs.DS↗

arXiv2019

Stochastic Online Metric Matching

We study the minimum-cost metric perfect matching problem under online i.i.d arrivals. We are given a fixed metric with a server at each of the points, and then requests arrive online, each drawn independently from a known probability distribution over the points. Each request has to be matched to a free server, with cost equal to the distance. The goal is to minimize the expected total cost of the matching. Such stochastic arrival models have been widely studied for the maximization variants of the online matching problem; however, the only known result for the minimization problem is a tight $O(\log n)$-competitiveness for the random-order arrival model. This is in contrast with the adversarial model, where an optimal competitive ratio of $O(\log n)$ has long been conjectured and remains a tantalizing open question. In this paper, we show improved results in the i.i.d arrival model. We show how the i.i.d model can be used to give substantially better algorithms: our main result is an $O((\log \log \log n)^2)$-competitive algorithm in this model. Along the way we give a $9$-competitive algorithm for the line and tree metrics. Both results imply a strict separation between the i.i.d model and the adversarial and random order models, both for general metrics and these much-studied metrics.

cs.DS↗

arXiv2018

Tight Competitive Ratios of Classic Matching Algorithms in the Fully Online Model

Huang et al.~(STOC 2018) introduced the fully online matching problem, a generalization of the classic online bipartite matching problem in that it allows all vertices to arrive online and considers general graphs. They showed that the ranking algorithm by Karp et al.~(STOC 1990) is strictly better than $0.5$-competitive and the problem is strictly harder than the online bipartite matching problem in that no algorithms can be $(1-1/e)$-competitive. This paper pins down two tight competitive ratios of classic algorithms for the fully online matching problem. For the fractional version of the problem, we show that a natural instantiation of the water-filling algorithm is $2-\sqrt{2} \approx 0.585$-competitive, together with a matching hardness result. Interestingly, our hardness result applies to arbitrary algorithms in the edge-arrival models of the online matching problem, improving the state-of-art $\frac{1}{1+\ln 2} \approx 0.5906$ upper bound. For integral algorithms, we show a tight competitive ratio of $\approx 0.567$ for the ranking algorithm on bipartite graphs, matching a hardness result by Huang et al. (STOC 2018).

cs.DS↗

arXiv2017

Reinforcement Mechanism Design, with Applications to Dynamic Pricing in Sponsored Search Auctions

In this study, we apply reinforcement learning techniques and propose what we call reinforcement mechanism design to tackle the dynamic pricing problem in sponsored search auctions. In contrast to previous game-theoretical approaches that heavily rely on rationality and common knowledge among the bidders, we take a data-driven approach, and try to learn, over repeated interactions, the set of optimal reserve prices. We implement our approach within the current sponsored search framework of a major search engine: we first train a buyer behavior model, via a real bidding data set, that accurately predicts bids given information that bidders are aware of, including the game parameters disclosed by the search engine, as well as the bidders' KPI data from previous rounds. We then put forward a reinforcement/MDP (Markov Decision Process) based algorithm that optimizes reserve prices over time, in a GSP-like auction. Our simulations demonstrate that our framework outperforms static optimization strategies including the ones that are currently in use, as well as several other dynamic ones.

cs.GT↗