arXiv research

A conformal procedure improves CoT reasoning by aggregating reasoning paths and calibrating abstention rules.

Temporal aggregation reveals latent default correlation from monthly data.

PathNNs improve graph neural networks by distinguishing non-isomorphic graphs.

Temporal coarse-graining of latent default paths explains effective correlation in corporate defaults.

ie-HGCN addresses HIN challenges by efficiently learning node representations.

Study online learning in MDPs with aggregate bandit feedback, achieving low regret in both stochastic and adversarial settings.

We describe the sample paths of the stochastic field $F = F_t(v,x,q)$ of aggregate utilities parameterized by Pareto weights $v$ and total cash amounts $x$ and stocks' quantities $q$ in an economy. We also describe the sample paths of the stochastic field $G = G_t(u,y,q)$, which is conjugate to $F$ with respect to the …

Graph neural network have achieved impressive results in predicting molecular properties, but they do not directly account for local and hidden structures in the graph such as functional groups and molecular geometry. At each propagation step, GNNs aggregate only over first order neighbours, ignoring important informat…

This research develops a new model for cyber risk and insurance pricing.

We provide a general construction of time-consistent sublinear expectations on the space of continuous paths. It yields the existence of the conditional G-expectation of a Borel-measurable (rather than quasi-continuous) random variable, a generalization of the random G-expectation, and an optional sampling theorem that…

Non-affine aggregation rules cannot preserve monotonicity in convex learning.

Recently, researchers have started decomposing deep neural network models according to their semantics or functions. Recent work has shown the effectiveness of decomposed functional blocks for defending adversarial attacks, which add small input perturbation to the input image to fool the DNN models. This work proposes…

Graph edges, along with their labels, can represent information of fundamental importance, such as links between web pages, friendship between users, the rating given by users to other users or items, and much more. We introduce LEAP, a trainable, general framework for predicting the presence and properties of edges on…

Much of the recent work on learning molecular representations has been based on Graph Convolution Networks (GCN). These models rely on local aggregation operations and can therefore miss higher-order graph properties. To remedy this, we propose Path-Augmented Graph Transformer Networks (PAGTN) that are explicitly built…

We provide an explicit aggregation in the neoclassical growth model with aggregate shocks and uninsurable employment risk. We show there are two restrictions on the unemployment shock for approximate aggregation to occur. First the probability of unemployment must be positive for each agent in each time period. That en…

CDA framework infers channel influence from aggregated data without user identifiers.

A model for insider trading with past price dependencies.

EntroPath learns manifold geometry from diffusion paths.

New method improves MAP inference for CGMs on path graphs, avoiding approximation and maintaining integrality.

MELO predicts electricity loads by adapting to shifts without external indicators.

GDP of China is about 11 trillion dollars and GDP of the United States is about 18 trillion dollars. Suppose that we know for the coming years, economy of the US will experience a real growth rate equal to \%3 and economy of China will experience a real growth as of \%6. Now, the question is how long does it take for e…

A new method for portfolio allocation in continuous-time markets.

Estimating temporal patterns in travel times along road segments in urban settings is of central importance to traffic engineers and city planners. In this work, we propose a methodology to leverage coarse-grained and aggregated travel time data to estimate the street-level travel times of a given metropolitan area. Ou…

A new graph kernel uses LCS and Wasserstein distance for better graph comparisons.

A new method for feature selection robust to noise and design variability.

A new network learns to prioritize messages for efficient multi-robot path planning.

A new method extracts events and their arguments efficiently from text.

CDP reduces point cloud dimensions by preserving detour-induced local non-convexity.

We address the problem of estimating the parameters of a time-homogeneous Markov chain given only noisy, aggregate data. This arises when a population of individuals behave independently according to a Markov chain, but individual sample paths cannot be observed due to limitations of the observation process or the need…

CAPM interpretation is flawed; beta reflects proxy for underlying driver, not causal transmission.

In this paper, we focus on graph representation learning of heterogeneous information network (HIN), in which various types of vertices are connected by various types of relations. Most of the existing methods conducted on HIN revise homogeneous graph embedding models via meta-paths to learn low-dimensional vector spac…

SGM combines deep learning and planning for robust long-horizon tasks.

The celebrated Sequence to Sequence learning (Seq2Seq) technique and its numerous variants achieve excellent performance on many tasks. However, many machine learning tasks have inputs naturally represented as graphs; existing Seq2Seq models face a significant challenge in achieving accurate conversion from graph form …

The length of the geodesic between two data points along a Riemannian manifold, induced by a deep generative model, yields a principled measure of similarity. Current approaches are limited to low-dimensional latent spaces, due to the computational complexity of solving a non-convex optimisation problem. We propose fin…

This paper models default data to capture dynamic dependence across sectors.

Recently, several studies have explored methods for using KG embedding to answer logical queries. These approaches either treat embedding learning and query answering as two separated learning tasks, or fail to deal with the variability of contributions from different query paths. We proposed to leverage a graph attent…

Temporal coarse-graining of multi-sector default count data generates effective correlation matrices and rank copulas.

This paper presents an analytical treatment of economic systems with an arbitrary number of agents that keeps track of the systems' interactions and agents' complexity. This formalism does not seek to aggregate agents. It rather replaces the standard optimization approach by a probabilistic description of both the enti…

MF-PID uses interacting samples to efficiently transport probability mass.

We present a neural model for representing snippets of code as continuous distributed vectors ("code embeddings"). The main idea is to represent a code snippet as a single fixed-length $\textit{code vector}$, which can be used to predict semantic properties of the snippet. This is performed by decomposing code to a col…

Proposes a new optimization-based method for aggregating sets in neural networks.

Graph algorithms are key tools in many fields of science and technology. Some of these algorithms depend on propagating information between distant nodes in a graph. Recently, there have been a number of deep learning architectures proposed to learn on undirected graphs. However, most of these architectures aggregate i…

Study aggregation of statistical evidence under unknown dependence using group-invariance.

New LCM aggregator improves GNN performance and efficiency.

Unified approach to aggregating models and preferences.

Federated learning is a distributed framework for training machine learning models over the data residing at mobile devices, while protecting the privacy of individual users. A major bottleneck in scaling federated learning to a large number of users is the overhead of secure model aggregation across many users. In par…

In order to scale standard Gaussian process (GP) regression to large-scale datasets, aggregation models employ factorized training process and then combine predictions from distributed experts. The state-of-the-art aggregation models, however, either provide inconsistent predictions or require time-consuming aggregatio…

Aggregation challenges causal interpretation of IV estimators.