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.
Alternative sampling method for autoregressive models using Langevin dynamics.
Improved flow-based models capture dependencies better with multi-scale autoregressive priors.
AR-Flow VAE improves blind source separation with flexible autoregressive priors.
New approach uses autoregressive models to explore and quantify uncertainty in decision-making.
Non-autoregressive method speeds up protein folding prediction 23 times.
A new model trains prior and encoder/decoder networks simultaneously for efficient generation.
Autoregressive generative models of images tend to be biased towards capturing local structure, and as a result they often produce samples which are lacking in terms of large-scale coherence. To address this, we propose two methods to learn discrete representations of images which abstract away local detail. We show th…
This paper compares log-likelihood and BLEU scores for sequence generation tasks.
New MIF architecture improves posterior approximations in Bayesian models.
Efficiently combines autoregressive and set-based models for joint distributions.
Paper proposes ARPHMM for fault detection and prognosis in aero-engines.
The pricing of financial derivatives, which requires massive calculations and close-to-real-time operations under many trading and arbitrage scenarios, were largely infeasible in the past. However, with the advancement of modern computing, the efficiency has substantially improved. In this work, we propose and design a…
SAHMM-VAE separates sources adaptively using hidden Markov priors.
Bayesian VAR model discovers Granger causality with uncertainty-aware binary graphs.
WaveGrad generates high-fidelity audio using gradient estimation.
We present the Insertion Transformer, an iterative, partially autoregressive model for sequence generation based on insertion operations. Unlike typical autoregressive models which rely on a fixed, often left-to-right ordering of the output, our approach accommodates arbitrary orderings by allowing for tokens to be ins…
Neural autoregressive models are explicit density estimators that achieve state-of-the-art likelihoods for generative modeling. The D-dimensional data distribution is factorized into an autoregressive product of one-dimensional conditional distributions according to the chain rule. Data completion is a more involved ta…
SHVC improves image compression with fewer parameters.
New method for density estimation without approximating posterior distributions.
Bayesian framework selects features and lags for time series forecasting.
Flow-based generative models are powerful exact likelihood models with efficient sampling and inference. Despite their computational efficiency, flow-based models generally have much worse density modeling performance compared to state-of-the-art autoregressive models. In this paper, we investigate and improve upon thr…
We explore the use of Vector Quantized Variational AutoEncoder (VQ-VAE) models for large scale image generation. To this end, we scale and enhance the autoregressive priors used in VQ-VAE to generate synthetic samples of much higher coherence and fidelity than possible before. We use simple feed-forward encoder and dec…
mGENRE improves multilingual entity linking with autoregressive sequence prediction.
LTMs use latent vectors for efficient autoregressive generation.
In this paper we present a a deep generative model for lossy video compression. We employ a model that consists of a 3D autoencoder with a discrete latent space and an autoregressive prior used for entropy coding. Both autoencoder and prior are trained jointly to minimize a rate-distortion loss, which is closely relate…
Neural networks improve gravitational-wave parameter estimation.
Bayesian model predicts interest rates with short-term accuracy and long-term stability.
We introduce a new criterion to determine the order of an autoregressive model fitted to time series data. It has the benefits of the two well-known model selection techniques, the Akaike information criterion and the Bayesian information criterion. When the data is generated from a finite order autoregression, the Bay…
Enhances FAVAR models with autoencoder for better economic forecasting and interpretability.
In this paper, we study two aspects of the variational autoencoder (VAE): the prior distribution over the latent variables and its corresponding posterior. First, we decompose the learning of VAEs into layerwise density estimation, and argue that having a flexible prior is beneficial to both sample generation and infer…
Many complex dynamical phenomena can be effectively modeled by a system that switches among a set of conditionally linear dynamical modes. We consider two such models: the switching linear dynamical system (SLDS) and the switching vector autoregressive (VAR) process. Our Bayesian nonparametric approach utilizes a hiera…
SIGMA prior enables federated learning for non-factorizable models.
SpinSVAR estimates SVAR models with sparse input, improving accuracy and scalability.
VQ-DRAW compresses images and generates realistic samples.
Recent neural text-to-speech (TTS) models with fine-grained latent features enable precise control of the prosody of synthesized speech. Such models typically incorporate a fine-grained variational autoencoder (VAE) structure, extracting latent features at each input token (e.g., phonemes). However, generating samples …
Autoregressive models are among the best performing neural density estimators. We describe an approach for increasing the flexibility of an autoregressive model, based on modelling the random numbers that the model uses internally when generating data. By constructing a stack of autoregressive models, each modelling th…
Develops effective adversarial attacks on probabilistic forecasting models.
Develops a new model for network estimation from multi-variate data.
This work proposes an efficient autoregressive model for text generation.
Representation learning seeks to expose certain aspects of observed data in a learned representation that's amenable to downstream tasks like classification. For instance, a good representation for 2D images might be one that describes only global structure and discards information about detailed texture. In this paper…
We present a windowed technique to learn parsimonious time-varying autoregressive models from multivariate timeseries. This unsupervised method uncovers interpretable spatiotemporal structure in data via non-smooth and non-convex optimization. In each time window, we assume the data follow a linear model parameterized …
Autoregressive sequence models achieve state-of-the-art performance in domains like machine translation. However, due to the autoregressive factorization nature, these models suffer from heavy latency during inference. Recently, non-autoregressive sequence models were proposed to reduce the inference time. However, the…
Bayesian method for multivariate autoregressive models with exogenous inputs.
Autoregressive state transitions, where predictions are conditioned on past predictions, are the predominant choice for both deterministic and stochastic sequential models. However, autoregressive feedback exposes the evolution of the hidden state trajectory to potential biases from well-known train-test discrepancies.…
We introduce a method for learning the dynamics of complex nonlinear systems based on deep generative models over temporal segments of states and actions. Unlike dynamics models that operate over individual discrete timesteps, we learn the distribution over future state trajectories conditioned on past state, past acti…
Self-reflective VAE improves inference and generative modeling without complex components.
Proposes DCNAR for dynamic causal inference from neural time series.
Paper proposes AXE loss for non-autoregressive machine translation, improving performance.