Searcharxiv⌕ Search

arXiv subjects

Yuying Li

Publications and source records attributed to Yuying Li.

31 records · Page 2Linked to original sources

The Adversarial AI-Art: Understanding, Generation, Detection, and Benchmarking

Generative AI models can produce high-quality images based on text prompts. The generated images often appear indistinguishable from images generated by conventional optical photography devices or created by human artists (i.e., real images). While the outstanding performance of such generative models is generally well received, security concerns arise. For instance, such image generators could be used to facilitate fraud or scam schemes, generate and spread misinformation, or produce fabricated artworks. In this paper, we present a systematic attempt at understanding and detecting AI-generated images (AI-art) in adversarial scenarios. First, we collect and share a dataset of real images and their corresponding artificial counterparts generated by four popular AI image generators. The dataset, named ARIA, contains over 140K images in five categories: artworks (painting), social media images, news photos, disaster scenes, and anime pictures. This dataset can be used as a foundation to support future research on adversarial AI-art. Next, we present a user study that employs the ARIA dataset to evaluate if real-world users can distinguish with or without reference images. In a benchmarking study, we further evaluate if state-of-the-art open-source and commercial AI image detectors can effectively identify the images in the ARIA dataset. Finally, we present a ResNet-50 classifier and evaluate its accuracy and transferability on the ARIA dataset.

cs.CV↗

An ethical study of generative AI from the Actor-Network Theory perspective

The widespread use of Generative Artificial Intelligence in the innovation and generation of communication content is mainly due to its exceptional creative ability, operational efficiency, and compatibility with diverse industries. Nevertheless, this has also sparked ethical problems, such as unauthorized access to data, biased decision-making by algorithms, and criminal use of generated content. In order to tackle the security vulnerabilities linked to Generative Artificial Intelligence, we analyze ChatGPT as a case study within the framework of Actor-Network Theory. We have discovered a total of nine actors, including both human and non-human creatures. We examine the actors and processes of translation involved in the ethical issues related to ChatGPT and analyze the key players involved in the emergence of moral issues. The objective is to explore the origins of the ethical issues that arise with Generative Artificial Intelligence and provide a particular perspective on the governance of Generative Artificial Intelligence.

cs.CY↗

Machine Learning and Hamilton-Jacobi-Bellman Equation for Optimal Decumulation: a Comparison Study

We propose a novel data-driven neural network (NN) optimization framework for solving an optimal stochastic control problem under stochastic constraints. Customized activation functions for the output layers of the NN are applied, which permits training via standard unconstrained optimization. The optimal solution yields a multi-period asset allocation and decumulation strategy for a holder of a defined contribution (DC) pension plan. The objective function of the optimal control problem is based on expected wealth withdrawn (EW) and expected shortfall (ES) that directly targets left-tail risk. The stochastic bound constraints enforce a guaranteed minimum withdrawal each year. We demonstrate that the data-driven approach is capable of learning a near-optimal solution by benchmarking it against the numerical results from a Hamilton-Jacobi-Bellman (HJB) Partial Differential Equation (PDE) computational framework.

math.OC↗

Neural Network Approach to Portfolio Optimization with Leverage Constraints:a Case Study on High Inflation Investment

Motivated by the current global high inflation scenario, we aim to discover a dynamic multi-period allocation strategy to optimally outperform a passive benchmark while adhering to a bounded leverage limit. To this end, we formulate an optimal control problem to outperform a benchmark portfolio throughout the investment horizon. Assuming the asset prices follow the jump-diffusion model during high inflation periods, we first establish a closed-form solution for the optimal strategy that outperforms a passive strategy under the cumulative quadratic tracking difference (CD) objective, assuming continuous trading and no bankruptcy. To obtain strategies under the bounded leverage constraint among other realistic constraints, we then propose a novel leverage-feasible neural network (LFNN) to represent control, which converts the original constrained optimization problem into an unconstrained optimization problem that is computationally feasible with standard optimization methods. We establish mathematically that the LFNN approximation can yield a solution that is arbitrarily close to the solution of the original optimal control problem with bounded leverage. We further apply the LFNN approach to a four-asset investment scenario with bootstrap resampled asset returns from the filtered high inflation regime data. The LFNN strategy is shown to consistently outperform the passive benchmark strategy by about 200 bps (median annualized return), with a greater than 90% probability of outperforming the benchmark at the end of the investment horizon.

q-fin.PM↗

A parsimonious neural network approach to solve portfolio optimization problems without using dynamic programming

We present a parsimonious neural network approach, which does not rely on dynamic programming techniques, to solve dynamic portfolio optimization problems subject to multiple investment constraints. The number of parameters of the (potentially deep) neural network remains independent of the number of portfolio rebalancing events, and in contrast to, for example, reinforcement learning, the approach avoids the computation of high-dimensional conditional expectations. As a result, the approach remains practical even when considering large numbers of underlying assets, long investment time horizons or very frequent rebalancing events. We prove convergence of the numerical solution to the theoretical optimal solution of a large class of problems under fairly general conditions, and present ground truth analyses for a number of popular formulations, including mean-variance and mean-conditional value-at-risk problems. We also show that it is feasible to solve Sortino ratio-inspired objectives (penalizing only the variance of wealth outcomes below the mean) in dynamic trading settings with the proposed approach. Using numerical experiments, we demonstrate that if the investment objective functional is separable in the sense of dynamic programming, the correct time-consistent optimal investment strategy is recovered, otherwise we obtain the correct pre-commitment (time-inconsistent) investment strategy. The proposed approach remains agnostic as to the underlying data generating assumptions, and results are illustrated using (i) parametric models for underlying asset returns, (ii) stationary block bootstrap resampling of empirical returns, and (iii) generative adversarial network (GAN)-generated synthetic asset returns.

q-fin.CP↗

A Composite T60 Regression and Classification Approach for Speech Dereverberation

Dereverberation is often performed directly on the reverberant audio signal, without knowledge of the acoustic environment. Reverberation time, T60, however, is an essential acoustic factor that reflects how reverberation may impact a signal. In this work, we propose to perform dereverberation while leveraging key acoustic information from the environment. More specifically, we develop a joint learning approach that uses a composite T60 module and a separate dereverberation module to simultaneously perform reverberation time estimation and dereverberation. The reverberation time module provides key features to the dereverberation module during fine tuning. We evaluate our approach in simulated and real environments, and compare against several approaches. The results show that this composite framework improves performance in environments.

eess.AS↗

Optimal Asset Allocation For Outperforming A Stochastic Benchmark Target

We propose a data-driven Neural Network (NN) optimization framework to determine the optimal multi-period dynamic asset allocation strategy for outperforming a general stochastic target. We formulate the problem as an optimal stochastic control with an asymmetric, distribution shaping, objective function. The proposed framework is illustrated with the asset allocation problem in the accumulation phase of a defined contribution pension plan, with the goal of achieving a higher terminal wealth than a stochastic benchmark. We demonstrate that the data-driven approach is capable of learning an adaptive asset allocation strategy directly from historical market returns, without assuming any parametric model of the financial market dynamics. Following the optimal adaptive strategy, investors can make allocation decisions simply depending on the current state of the portfolio. The optimal adaptive strategy outperforms the benchmark constant proportion strategy, achieving a higher terminal wealth with a 90% probability, a 46% higher median terminal wealth, and a significantly more right-skewed terminal wealth distribution. We further demonstrate the robustness of the optimal adaptive strategy by testing the performance of the strategy on bootstrap resampled market data, which has different distributions compared to the training data.

q-fin.CP↗

On the index of unbalanced signed bicyclic graphs

In this paper, we focus on the index ( largest eigenvalue) of the adjacency matrix of connected signed graphs. We give some general results on the index when the corresponding signed graph is perturbed. As applications, we determine the first five largest index among all unbalanced bicyclic graphs on n >= 36 vertices together with the corresponding extremal signed graphs whose index attain these values.

math.CO↗

Quantum walk inspired algorithm for graph similarity and isomorphism

Large scale complex systems, such as social networks, electrical power grid, database structure, consumption pattern or brain connectivity, are often modeled using network graphs. Valuable insight can be gained by measuring the similarity between network graphs in order to make quantitative comparisons. Since these networks can be very large, scalability and efficiency of the algorithm are key concerns. More importantly, for graphs with unknown labeling, this graph similarity problem requires exponential time to solve using existing algorithms. In this paper, we propose a quantum walk inspired algorithm, which provides a solution to the graph similarity problem without prior knowledge on graph labeling. This algorithm is capable of distinguishing between minor structural differences, such as between strongly regular graphs with the same parameters. The algorithm has polynomial complexity, scaling with $O(n^9)$.

quant-ph↗

EREL Selection using Morphological Relation

This work concentrates on Extremal Regions of Extremum Level (EREL) selection. EREL is a recently proposed feature detector aiming at detecting regions from a set of extremal regions. This is a branching problem derived from segmentation of arterial wall boundaries from Intravascular Ultrasound (IVUS) images. For each IVUS frame, a set of EREL regions is generated to describe the luminal area of human coronary. Each EREL is then fitted by an ellipse to represent the luminal border. The goal is to assign the most appropriate EREL as the lumen. In this work, EREL selection carries out in two rounds. In the first round, the pattern in a set of EREL regions is analyzed and used to generate an approximate luminal region. Then, the two-dimensional (2D) correlation coefficients are computed between this approximate region and each EREL to keep the ones with tightest relevance. In the second round, a compactness measure is calculated for each EREL and its fitted ellipse to guarantee that the resulting EREL has not affected by the common artifacts such as bifurcations, shadows, and side branches. We evaluated the selected ERELs in terms of Hausdorff Distance (HD) and Jaccard Measure (JM) on the train and test set of a publicly available dataset. The results show that our selection strategy outperforms the current state-of-the-art.

cs.CV↗

Nonsmooth Frank-Wolfe using Uniform Affine Approximations

Frank-Wolfe methods (FW) have gained significant interest in the machine learning community due to its ability to efficiently solve large problems that admit a sparse structure (e.g. sparse vectors and low-rank matrices). However the performance of the existing FW method hinges on the quality of the linear approximation. This typically restricts FW to smooth functions for which the approximation quality, indicated by a global curvature measure, is reasonably good. In this paper, we propose a modified FW algorithm amenable to nonsmooth functions by optimizing for approximation quality over all affine approximations given a neighborhood of interest. We analyze theoretical properties of the proposed algorithm and demonstrate that it overcomes many issues associated with existing methods in the context of nonsmooth low-rank matrix estimation.

stat.ML↗

Projection Free Rank-Drop Steps

The Frank-Wolfe (FW) algorithm has been widely used in solving nuclear norm constrained problems, since it does not require projections. However, FW often yields high rank intermediate iterates, which can be very expensive in time and space costs for large problems. To address this issue, we propose a rank-drop method for nuclear norm constrained problems. The goal is to generate descent steps that lead to rank decreases, maintaining low-rank solutions throughout the algorithm. Moreover, the optimization problems are constrained to ensure that the rank-drop step is also feasible and can be readily incorporated into a projection-free minimization method, e.g., Frank-Wolfe. We demonstrate that by incorporating rank-drop steps into the Frank-Wolfe algorithm, the rank of the solution is greatly reduced compared to the original Frank-Wolfe or its common variants.

stat.ML↗

Are target date funds dinosaurs? Failure to adapt can lead to extinction

Investors in Target Date Funds are automatically switched from high risk to low risk assets as their retirements approach. Such funds have become very popular, but our analysis brings into question the rationale for them. Based on both a model with parameters fitted to historical returns and on bootstrap resampling, we find that adaptive investment strategies significantly outperform typical Target Date Fund strategies. This suggests that the vast majority of Target Date Funds are serving investors poorly.

q-fin.PM↗