A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
A new method predicts student skill success rates in real-time.
New method forecasts workforce reintegration success rates.
Machine learning predicts TV show success based on factors like characters and direction.
The muti-layer information bottleneck (IB) problem, where information is propagated (or successively refined) from layer to layer, is considered. Based on information forwarded by the preceding layer, each stage of the network is required to preserve a certain level of relevance with regards to a specific hidden variab…
Proposes a new framework to manage venture capital portfolio risk by focusing on deal-level correlations.
System recommends workouts and predicts success rates using RNNs.
GAIL with neural networks converges to global optima and has a known rate.
In this paper, we study the sum rate maximization for successive zero-forcing dirty-paper coding (SZFDPC) with per-antenna power constraint (PAPC). Although SZFDPC is a low-complexity alternative to the optimal dirty paper coding (DPC), efficient algorithms to compute its sum rate are still open problems especially und…
The state-of-the-art performance of deep learning algorithms has led to a considerable increase in the utilization of machine learning in security-sensitive and critical applications. However, it has recently been shown that a small and carefully crafted perturbation in the input space can completely fool a deep model.…
This paper studies a class of optimal multiple stopping problems driven by Lévy processes. Our model allows for a negative effective discount rate, which arises in a number of financial applications, including stock loans and real options, where the strike price can potentially grow at a higher rate than the original d…
Deep neural networks (DNNs) are known vulnerable to adversarial attacks. That is, adversarial examples, obtained by adding delicately crafted distortions onto original legal inputs, can mislead a DNN to classify them as any target labels. This work provides a solution to hardening DNNs under adversarial attacks through…
Deep neural networks outperform traditional methods in high-dimensional classification.
The paper analyzes how momentum affects convergence in stochastic gradient methods.
Success conditioning optimizes policies by imitating successful trajectories, solving a trust-region optimization problem.
Deep 3D models are vulnerable to isometry transformations under adversarial attacks.
Paper presents a defense framework against adversarial examples.
The convergence rate and final performance of common deep learning models have significantly benefited from heuristics such as learning rate schedules, knowledge distillation, skip connections, and normalization layers. In the absence of theoretical underpinnings, controlled experiments aimed at explaining these strate…
Numerous methods for crafting adversarial examples were proposed recently with high success rate. Since most existing machine learning based classifiers normalize images into some continuous, real vector, domain firstly, attacks often craft adversarial examples in such domain. However, "adversarial" examples may become…
AdversarialPSO uses fewer queries to create high-success-rate adversarial examples.
New learning rate schedule improves deep learning performance.
The paper provides convergence guarantees for multicalibration gradient boosting.
Paper establishes convergence rates and concentration bounds for stochastic approximation and reinforcement learning with Markovian noise.
AI-GAN generates realistic adversarial examples efficiently.
Tool manipulation is vital for facilitating robots to complete challenging task goals. It requires reasoning about the desired effect of the task and thus properly grasping and manipulating the tool to achieve the task. Task-agnostic grasping optimizes for grasp robustness while ignoring crucial task-specific constrain…
The paper analyzes RLVR's training dynamics, proving convergence depends on aligning update direction with Gradient Gap.
Recent work has established an empirically successful framework for adapting learning rates for stochastic gradient descent (SGD). This effectively removes all needs for tuning, while automatically reducing learning rates over time on stationary problems, and permitting learning rates to grow appropriately in non-stati…
New convergence rates for shuffling gradient methods without strong convexity.
TREK uses distillation to help students solve hard problems.
Proposes a method to select features for deep learning in noisy, high-dimensional data.
Vanguard uses AI to create personalized financial plans.
E-valuator converts verifier scores into reliable decision rules.
Despite the great empirical success of deep reinforcement learning, its theoretical foundation is less well understood. In this work, we make the first attempt to theoretically understand the deep Q-network (DQN) algorithm (Mnih et al., 2015) from both algorithmic and statistical perspectives. In specific, we focus on …
Cyclical learning rate improves neural machine translation performance.
In many professons employees are rewarded according to their relative performance. Corresponding economy can be modeled by taking independent agents who gain from the market with a rate which depends on their current gain. We argue that this simple realistic rate generates a scale free distribution even though intr…
Active management is a term that has many meanings and we have found the defining characteristics needed for success as an "active manager" elusive within the literature. In this paper we offer a set of criteria that defines an active manager and his success. In order to facilitate this, we introduce several definition…
Large learning rates enhance model robustness and compressibility.
The Dirichlet random walk on manifolds has a positive escape rate if the cover is non-amenable.
We describe and analyze a new boosting algorithm for deep learning called SelfieBoost. Unlike other boosting algorithms, like AdaBoost, which construct ensembles of classifiers, SelfieBoost boosts the accuracy of a single network. We prove a convergence rate for SelfieBoost under some "SGD success" assumpti…
The paper introduces SuccessProbaMax to optimize policy success probability in online advertising.
The learning rate warmup heuristic achieves remarkable success in stabilizing training, accelerating convergence and improving generalization for adaptive stochastic optimization algorithms like RMSprop and Adam. Here, we study its mechanism in details. Pursuing the theory behind warmup, we identify a problem of the ad…
We present a novel optimization method, named the Combined Optimization Method (COM), for the joint optimization of two or more cost functions. Unlike the conventional joint optimization schemes, which try to find minima in a weighted sum of cost functions, the COM explores search space for common minima shared by all …
We consider the black-box adversarial setting, where the adversary has to generate adversarial perturbations without access to the target models to compute gradients. Previous methods tried to approximate the gradient either by using a transfer gradient of a surrogate white-box model, or based on the query feedback. Ho…
Defense against DL-based lithographic hotspot detectors backdooring attacks reduces success rate from 84% to ~0%
Inspired by the unsupervised learning or self-organization in the machine learning context, here we attempt to draw `learning curve' for the collective behavior of job-seeking `zero-intelligence' labors in successive job-hunting processes. Our labor market is supposed to be opened especially for university graduates in…
The aim of this work is to provide fast and accurate approximation schemes for the Monte-Carlo pricing of derivatives in the Lévy LIBOR model of Eberlein and Özkan (2005). Standard methods can be applied to solve the stochastic differential equations of the successive LIBOR rates but the methods are generally slow. We …
Sparse-RS framework efficiently attacks models with sparse perturbations.
Graph attacks can be successful with just a few bad nodes.
Stochastic gradient algorithms have been the main focus of large-scale learning problems and they led to important successes in machine learning. The convergence of SGD depends on the careful choice of learning rate and the amount of the noise in stochastic estimates of the gradients. In this paper, we propose a new ad…