A note on Kollár valuations
We prove the set of Kollár valuations in the dual complex of a klt singularity with a fixed complement is path connected. We also classify the case when the dual complex is one dimensional.
math.AG↗
arXiv subjects
Chenyang Xu
Publications and source records attributed to Chenyang Xu.
Fire in SRRN: Next-Gen 3D Temperature Field Reconstruction Technology
In aerospace and energy engineering, accurate 3D combustion field temperature measurement is critical. The resolution of traditional methods based on algebraic iteration is limited by the initial voxel division. This study introduces a novel method for reconstructing three-dimensional temperature fields using the Spatial Radiation Representation Network (SRRN). This method utilizes the flame thermal radiation characteristics and differentiable rendering in graphics, and combines it with a multi-layer perceptron to achieve a functional representation of the flame temperature field. The effectiveness of SRRN is evaluated through simulated temperature field reconstruction experiments with different levels of complexity. The maximum root mean square error is 10.17, which proves the robustness of the algorithm to Gaussian noise and salt-and-pepper noise. We conducted a butane flame temperature field reconstruction experiment, and the maximum relative error between the reconstruction result and the thermocouple measurement value was 4.86%, confirming that the algorithm can achieve accurate reconstruction.
eess.IV↗
K-stability for varieties with a big anticanonical class
We extend the algebraic K-stability theory to projective klt pairs with a big anticanonical class. While in general such a pair could behave pathologically, it is observed in this note that K-semistability condition will force them to have a klt anticanonical model, whose stability property is the same as the original pair.
math.AG↗
A Clustering-guided Contrastive Fusion for Multi-view Representation Learning
The past two decades have seen increasingly rapid advances in the field of multi-view representation learning due to it extracting useful information from diverse domains to facilitate the development of multi-view applications. However, the community faces two challenges: i) how to learn robust representations from a large amount of unlabeled data to against noise or incomplete views setting, and ii) how to balance view consistency and complementary for various downstream tasks. To this end, we utilize a deep fusion network to fuse view-specific representations into the view-common representation, extracting high-level semantics for obtaining robust representation. In addition, we employ a clustering task to guide the fusion network to prevent it from leading to trivial solutions. For balancing consistency and complementary, then, we design an asymmetrical contrastive strategy that aligns the view-common representation and each view-specific representation. These modules are incorporated into a unified method known as CLustering-guided cOntrastiVE fusioN (CLOVEN). We quantitatively and qualitatively evaluate the proposed method on five datasets, demonstrating that CLOVEN outperforms 11 competitive multi-view learning methods in clustering and classification. In the incomplete view scenario, our proposed method resists noise interference better than those of our competitors. Furthermore, the visualization analysis shows that CLOVEN can preserve the intrinsic structure of view-specific representation while also improving the compactness of view-commom representation. Our source code will be available soon at https://github.com/guanzhou-ke/cloven.
cs.CV↗
Disentangling Multi-view Representations Beyond Inductive Bias
Multi-view (or -modality) representation learning aims to understand the relationships between different view representations. Existing methods disentangle multi-view representations into consistent and view-specific representations by introducing strong inductive biases, which can limit their generalization ability. In this paper, we propose a novel multi-view representation disentangling method that aims to go beyond inductive biases, ensuring both interpretability and generalizability of the resulting representations. Our method is based on the observation that discovering multi-view consistency in advance can determine the disentangling information boundary, leading to a decoupled learning objective. We also found that the consistency can be easily extracted by maximizing the transformation invariance and clustering consistency between views. These observations drive us to propose a two-stage framework. In the first stage, we obtain multi-view consistency by training a consistent encoder to produce semantically-consistent representations across views as well as their corresponding pseudo-labels. In the second stage, we disentangle specificity from comprehensive representations by minimizing the upper bound of mutual information between consistent and comprehensive representations. Finally, we reconstruct the original data by concatenating pseudo-labels and view-specific representations. Our experiments on four multi-view datasets demonstrate that our proposed method outperforms 12 comparison methods in terms of clustering and classification performance. The visualization results also show that the extracted consistency and specificity are compact and interpretable. Our code can be found at \url{https://github.com/Guanzhou-Ke/DMRIB}.
cs.CV↗
Auction Design for Value Maximizers with Budget and Return-on-spend Constraints
The paper designs revenue-maximizing auction mechanisms for agents who aim to maximize their total obtained values rather than the classical quasi-linear utilities. Several models have been proposed to capture the behaviors of such agents in the literature. In the paper, we consider the model where agents are subject to budget and return-on-spend constraints. The budget constraint of an agent limits the maximum payment she can afford, while the return-on-spend constraint means that the ratio of the total obtained value (return) to the total payment (spend) cannot be lower than the targeted bar set by the agent. The problem was first coined by [Balseiro et al., EC 2022]. In their work, only Bayesian mechanisms were considered. We initiate the study of the problem in the worst-case model and compare the revenue of our mechanisms to an offline optimal solution, the most ambitious benchmark. The paper distinguishes two main auction settings based on the accessibility of agents' information: fully private and partially private. In the fully private setting, an agent's valuation, budget, and target bar are all private. We show that if agents are unit-demand, constant approximation mechanisms can be obtained; while for additive agents, there exists a mechanism that achieves a constant approximation ratio under a large market assumption. The partially private setting is the setting considered in the previous work [Balseiro et al., EC 2022] where only the agents' target bars are private. We show that in this setting, the approximation ratio of the single-item auction can be further improved, and a $Ω(1/\sqrt{n})$-approximation mechanism can be derived for additive agents.
cs.GT↗
Motivic zeta function via dlt modification
Given a smooth variety $X$ and a regular function $f$ on it, by considering the dlt modification, we define the dlt motivic zeta function $Z^{\rm dlt}_{\rm mot}(s)$ which does not depend on the choice of the dlt modification.
math.AG↗
Scheduling with a Limited Testing Budget
Scheduling with testing falls under the umbrella of the research on optimization with explorable uncertainty. In this model, each job has an upper limit on its processing time that can be decreased to a lower limit (possibly unknown) by some preliminary action (testing). Recently, D{ü}rr et al. \cite{DBLP:journals/algorithmica/DurrEMM20} has studied a setting where testing a job takes a unit time, and the goal is to minimize total completion time or makespan on a single machine. In this paper, we extend their problem to the budget setting in which each test consumes a job-specific cost, and we require that the total testing cost cannot exceed a given budget. We consider the offline variant (the lower processing time is known) and the oblivious variant (the lower processing time is unknown) and aim to minimize the total completion time or makespan on a single machine. For the total completion time objective, we show NP-hardness and derive a PTAS for the offline variant based on a novel LP rounding scheme. We give a $(4+ε)$-competitive algorithm for the oblivious variant based on a framework inspired by the worst-case lower-bound instance. For the makespan objective, we give an FPTAS for the offline variant and a $(2+ε)$-competitive algorithm for the oblivious variant. Our algorithms for the oblivious variants under both objectives run in time $O(poly(n/ε))$. Lastly, we show that our results are essentially optimal by providing matching lower bounds.
cs.DS↗
The existence of the Kähler-Ricci soliton degeneration
We prove an algebraic version of the Hamilton-Tian Conjecture for all log Fano pairs. More precisely, we show that any log Fano pair admits a canonical two-step degeneration to a reduced uniformly Ding stable triple, which admits a Kähler-Ricci soliton when the ground field $\mathbb{k}=\mathbb{C}$.
math.AG↗
Online Dynamic Acknowledgement with Learned Predictions
We revisit the online dynamic acknowledgment problem. In the problem, a sequence of requests arrive over time to be acknowledged, and all outstanding requests can be satisfied simultaneously by one acknowledgement. The goal of the problem is to minimize the total request delay plus acknowledgement cost. This elegant model studies the trade-off between acknowledgement cost and waiting experienced by requests. The problem has been well studied and the tight competitive ratios have been determined. For this well-studied problem, we focus on how to effectively use machine-learned predictions to have better performance. We develop algorithms that perform arbitrarily close to the optimum with accurate predictions while concurrently having the guarantees arbitrarily close to what the best online algorithms can offer without access to predictions, thereby achieving simultaneous optimum consistency and robustness. This new result is enabled by our novel prediction error measure. No error measure was defined for the problem prior to our work, and natural measures failed due to the challenge that requests with different arrival times have different effects on the objective. We hope our ideas can be used for other online problems with temporal aspects that have been resisting proper error measures.
cs.DS↗
Min-max Submodular Ranking for Multiple Agents
In the submodular ranking (SR) problem, the input consists of a set of submodular functions defined on a ground set of elements. The goal is to order elements for all the functions to have value above a certain threshold as soon on average as possible, assuming we choose one element per time. The problem is flexible enough to capture various applications in machine learning, including decision trees. This paper considers the min-max version of SR where multiple instances share the ground set. With the view of each instance being associated with an agent, the min-max problem is to order the common elements to minimize the maximum objective of all agents -- thus, finding a fair solution for all agents. We give approximation algorithms for this problem and demonstrate their effectiveness in the application of finding a decision tree for multiple agents.
cs.DS↗
Learning-Augmented Algorithms for Online Steiner Tree
This paper considers the recently popular beyond-worst-case algorithm analysis model which integrates machine-learned predictions with online algorithm design. We consider the online Steiner tree problem in this model for both directed and undirected graphs. Steiner tree is known to have strong lower bounds in the online setting and any algorithm's worst-case guarantee is far from desirable. This paper considers algorithms that predict which terminal arrives online. The predictions may be incorrect and the algorithms' performance is parameterized by the number of incorrectly predicted terminals. These guarantees ensure that algorithms break through the online lower bounds with good predictions and the competitive ratio gracefully degrades as the prediction error grows. We then observe that the theory is predictive of what will occur empirically. We show on graphs where terminals are drawn from a distribution, the new online algorithms have strong performance even with modestly correct predictions.
cs.LG↗
Mechanism Design with Predictions
Improving algorithms via predictions is a very active research topic in recent years. This paper initiates the systematic study of mechanism design in this model. In a number of well-studied mechanism design settings, we make use of imperfect predictions to design mechanisms that perform much better than traditional mechanisms if the predictions are accurate (consistency), while always retaining worst-case guarantees even with very imprecise predictions (robustness). Furthermore, we refer to the largest prediction error sufficient to give a good performance as the error tolerance of a mechanism, and observe that an intrinsic tradeoff among consistency, robustness and error tolerance is common for mechanism design with predictions.
cs.GT↗
Multiagent MST Cover: Pleasing All Optimally via A Simple Voting Rule
Given a connected graph on whose edges we can build roads to connect the nodes, a number of agents hold possibly different perspectives on which edges should be selected by assigning different edge weights. Our task is to build a minimum number of roads so that every agent has a spanning tree in the built subgraph whose weight is the same as a minimum spanning tree in the original graph. We first show that this problem is NP-hard and does not admit better than $((1-o(1))\ln k)$-approximation polynomial-time algorithms unless P=NP, where $k$ is the number of agents. We then give a simple voting algorithm with an optimal approximation ratio. Moreover, our algorithm only needs to access the agents' rankings on the edges. Finally, we extend our results to submodular objective functions and Matroid rank constraints.
cs.GT↗
Finite generation for valuations computing stability thresholds and applications to K-stability
We prove that on any log Fano pair of dimension $n$ whose stability threshold is less than $\frac{n+1}{n}$, any valuation computing the stability threshold has a finitely generated associated graded ring. Together with earlier works, this implies: (a) a log Fano pair is uniformly K-stable (resp. reduced uniformly K-stable) if and only if it is K-stable (resp. K-polystable); (b) the K-moduli spaces are proper and projective; and combining with the previously known equivalence between the existence of Kähler-Einstein metric and reduced uniform K-stability proved by the variational approach, (c) the Yau-Tian-Donaldson conjecture holds for general (possibly singular) log Fano pairs.
math.AG↗
Openness of K-semistability for Fano varieties
In this paper, we prove the openness of K-semistability in families of log Fano pairs by showing that the stability threshold is a constructible function on the fibers. We also prove that any special test configuration arises from a log canonical place of a bounded complement and establish properties of any minimizer of the stability threshold.
math.AG↗
On properness of K-moduli spaces and optimal degenerations of Fano varieties
We establish an algebraic approach to prove the properness of moduli spaces of K-polystable Fano varieties and reduce the problem to a conjecture on destabilizations of K-unstable Fano varieties. Specifically, we prove that if the stability threshold of every K-unstable Fano variety is computed by a divisorial valuation, then such K-moduli spaces are proper. The argument relies on studying certain optimal destabilizing test configurations and constructing a Theta-stratification on the moduli stack of Fano varieties.
math.AG↗
Learnable and Instance-Robust Predictions for Online Matching, Flows and Load Balancing
We propose a new model for augmenting algorithms with predictions by requiring that they are formally learnable and instance robust. Learnability ensures that predictions can be efficiently constructed from a reasonable amount of past data. Instance robustness ensures that the prediction is robust to modest changes in the problem input, where the measure of the change may be problem specific. Instance robustness insists on a smooth degradation in performance as a function of the change. Ideally, the performance is never worse than worst-case bounds. This also allows predictions to be objectively compared. We design online algorithms with predictions for a network flow allocation problem and restricted assignment makespan minimization. For both problems, two key properties are established: high quality predictions can be learned from a small sample of prior instances and these predictions are robust to errors that smoothly degrade as the underlying problem instance changes.
cs.LG↗