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.
Adaptive algorithm identifies best arm with abstention, showing phase transition from polynomial to exponential error probability.
Backdoors in deep neural networks are undetectable and enable invariance-based adversarial examples.
We inject undetectable backdoors into obfuscated neural networks and language models.
A time-dependent SIR model predicts COVID-19 spread and herd immunity.
For any positive integer r, we exhibit a knot Kr with (20 2 r--1 + 1) crossings whose Jones polynomial V (Kr) is equal to 1 mod-ulo 2 r. Our construction rests on a certain 20-crossing tangle T 20 which is undetectable by the Kauffman bracket polynomial pair mod 2.
Paper optimizes change-point detection using learned distributions from training sequences.
Adversarial examples raise questions about whether neural network models are sensitive to the same visual features as humans. In this paper, we first detect adversarial examples or otherwise corrupted images based on a class-conditional reconstruction of the input. To specifically attack our detection mechanism, we pro…
Intratumor heterogeneity is often manifested by vascular compartments with distinct pharmacokinetics that cannot be resolved directly by in vivo dynamic imaging. We developed tissue-specific compartment modeling (TSCM), an unsupervised computational method of deconvolving dynamic imaging series from heterogeneous tumor…
New method predicts positive samples with missing labels.
An Ensemble Anomaly Detection Framework for Risk Calculation Integrity
Closed essential surfaces in a three-manifold can be detected by ideal points of the character variety or by algebraic non-integral representations. We give examples of closed essential surfaces not detected in either of these ways. For ideal points, we use Chesebro's module-theoretic interpretation of Culler-Shalen th…
We present a deep learning system for testing graphics units by detecting novel visual corruptions in videos. Unlike previous work in which manual tagging was required to collect labeled training data, our weak supervision method is fully automatic and needs no human labelling. This is achieved by reproducing driver bu…
Undetected overfitting can occur when there are significant redundancies between training and validation data. We describe AVE, a new measure of training-validation redundancy for ligand-based classification problems that accounts for the similarity amongst inactive molecules as well as active. We investigated seven wi…
New framework detects time-varying economic persistence.
Kishino's knot is not detected by the fundamental group or the bracket polynomial; these invariants cannot differentiate between Kishino's knot and the unknot. However, we can show that Kishino's knot is not equivalent to unknot by applying either the 3-strand bracket polynomial or the surface bracket polynomial. In th…
Interview study reveals considerations for designing semi-automated bias detection tools.
Recent work establishes dataset difficulty and removes annotation artifacts via partial-input baselines (e.g., hypothesis-only models for SNLI or question-only models for VQA). When a partial-input baseline gets high accuracy, a dataset is cheatable. However, the converse is not necessarily true: the failure of a parti…
The study uses DCC for financial market analysis, revealing hidden correlations.
There has been an ongoing cycle where stronger defenses against adversarial attacks are subsequently broken by a more advanced defense-aware attack. We present a new approach towards ending this cycle where we "deflect'' adversarial attacks by causing the attacker to produce an input that semantically resembles the att…
New method separates graph structure from node attributes to recover lost signal.
Distributed learning is central for large-scale training of deep-learning models. However, they are exposed to a security threat in which Byzantine participants can interrupt or control the learning process. Previous attack models and their corresponding defenses assume that the rogue participants are (a) omniscient (k…
Deep Neural Networks (DNNs) are universal function approximators providing state-of- the-art solutions on wide range of applications. Common perceptual tasks such as speech recognition, image classification, and object tracking are now commonly tackled via DNNs. Some fundamental problems remain: (1) the lack of a mathe…
Artificial neural network (ANN) provides superior accuracy for nonlinear alternating current (AC) state estimation (SE) in smart grid over traditional methods. However, research has discovered that ANN could be easily fooled by adversarial examples. In this paper, we initiate a new study of adversarial false data injec…
Identification of patterns from discrete data time-series for statistical inference, threat detection, social opinion dynamics, brain activity prediction has received recent momentum. In addition to the huge data size, the associated challenges are, for example, (i) missing data to construct a closed time-varying compl…
Machine learning has begun to play a central role in many applications. A multitude of these applications typically also involve datasets that are distributed across multiple computing devices/machines due to either design constraints (e.g., multiagent systems) or computational/privacy reasons (e.g., learning on smartp…
New knot polynomials reveal patterns and mutations.
The financial crisis clearly illustrated the importance of characterizing the level of 'systemic' risk associated with an entire credit network, rather than with single institutions. However, the interplay between financial distress and topological changes is still poorly understood. Here we analyze the quarterly inter…
Deep neural networks have been demonstrated to be vulnerable to backdoor attacks. Specifically, by injecting a small number of maliciously constructed inputs into the training set, an adversary is able to plant a backdoor into the trained model. This backdoor can then be activated during inference by a backdoor trigger…
In this paper we study embeddings of oriented connected closed surfaces in . We define a complete invariant, the fundamental span, for such embeddings, generalizing the notion of the peripheral system of a knot group. From the fundamental span, several computable invariants are derived and employed to stud…
A new, efficient -modes algorithm improves clustering of categorical data.
ISAAC audits deep models for drug-target interactions, revealing structural differences.
This paper studies the equilibrium pricing of asset shares in the presence of dynamic private information. The market consists of a risk-neutral informed agent who observes the firm value, noise traders, and competitive market makers who set share prices using the total order flow as a noisy signal of the insider's inf…
New framework tackles fairness in link prediction beyond demographic parity.
New method targets relative risk heterogeneity in clinical trials.
Network operators are generally aware of common attack vectors that they defend against. For most networks the vast majority of traffic is legitimate. However new attack vectors are continually designed and attempted by bad actors which bypass detection and go unnoticed due to low volume. One strategy for finding such …
New method compares DP mechanisms beyond pairs.
Paper uses robust GMM to reconstruct missing data in Sentinel-2 images for crop monitoring.
New method allocates capital based on tail central moments for financial risk assessment.
We tackle anomaly detection in sparse time series data.
Knowledge about frequency and location of snow avalanche activity is essential for forecasting and mapping of snow avalanche hazard. Traditional field monitoring of avalanche activity has limitations, especially when surveying large and remote areas. In recent years, avalanche detection in Sentinel-1 radar satellite im…
This paper optimizes power grid protection settings to maximize network degradation due to cascading attacks.
We consider a three-layer Sejnowski machine and show that features learnt via contrastive divergence have a dual representation as patterns in a dense associative memory of order P=4. The latter is known to be able to Hebbian-store an amount of patterns scaling as N^{P-1}, where N denotes the number of constituting bin…
ODBAE detects complex phenotypes in biological data.
Study of participating policies with guaranteed minimum interest rate and surrender option.
The wave trace of certain convex domains can be smooth near some points in the length spectrum.
New method learns frequency-dependent partial correlations.
Perinatal stroke (PS) is a serious condition that, if undetected and thus untreated, often leads to life-long disability, in particular Cerebral Palsy (CP). In clinical settings, Prechtl's General Movement Assessment (GMA) can be used to classify infant movements using a Gestalt approach, identifying infants at high ri…
Machine learning improves diagnostic test accuracy for bovine tuberculosis.