22 papers · ranked by Valyu relevance
Roy Shivam Ram Shreshtth, Arnab Hazra, Gourab Mukherjee
Since the turn of the century, approximate Bayesian inference has steadily evolved as new computational techniques have been incorporated to handle increasingly complex and large-scale predictive problems. The recent success of deep neural networks and foundation models has now given rise to a new paradigm in…
Ishita Dasgupta, Eric Schulz, Noah D. Goodman, Samuel J. Gershman
Bayesian models of cognition assume that people compute probability distributions over hypotheses. However, the required computations are frequently intractable or prohibitively expensive. Since people often encounter many closely related distributions, selective reuse of computations (amortized inference) is a…
Joohwan Ko, Justin Domke
Amortized inference promises fast test-time Bayesian inference, but existing methods are inherently tied to fixed models. Extending amortization to unseen models typically requires retraining or costly test-time finetuning. In this paper, we ask: is it possible to build a single inference network capable of…
Daniel Ritchie, Paul Horsfall, Noah D. Goodman
Probabilistic programming languages (PPLs) are a powerful modeling tool, able to represent any computable probability distribution. Unfortunately, probabilistic program inference is often intractable, and existing PPLs mostly rely on expensive, approximate sampling-based methods. To alleviate this problem, one could…
Ishita Dasgupta, Eric Schulz, Noah D. Goodman, Samuel J. Gershman
Bayesian models of cognition posit that people compute probability distributions over hypotheses, possibly by constructing a sample-based approximation. Since people encounter many closely related distributions, a computationally efficient strategy is to selectively reuse computations – either the samples themselves or…
Rui Shu, Hung Bui, Shengjia Zhao, Mykel J. Kochenderfer + 1 more
'Stefano Ermon'] The variational autoencoder (VAE) is a popular model for density estimation and representation learning. Canonically, the variational principle suggests to prefer an expressive inference model so that the variational approximation is accurate. However, it is often overlooked that an overly-expressive…
Qasim Ramzan, Muhammad Amin, Shuhrah Alghamdi, Randa Alharbi + 1 more
The QR distribution, recently introduced for modeling lifetime data under Type-II censoring, offers a flexible framework for survival and reliability analysis. This study provides the first comprehensive evaluation of multiple modern estimation techniques for the QR distribution under Type-II censoring. We…
Lars Hertel, Neil Daftary, Fedor Borisyuk, Aman Gupta + 1 more
learning recommendation models Authors: ['Lars Hertel' 'Neil Daftary' 'Fedor Borisyuk' 'Aman Gupta' 'Rahul Mazumder'] We study user history modeling via Transformer encoders in deep learning recommendation models (DLRM). Such architectures can significantly improve recommendation quality, but usually incur high latency…
Alexander Fengler, Lakshmi N Govindarajan, Tony Chen, Michael J Frank + 2 more
In cognitive neuroscience, computational modeling can formally adjudicate between theories and affords quantitative fits to behavioral/brain data. Pragmatically, however, the space of plausible generative models considered is dramatically limited by the set of models with known likelihood functions. For many models…
Yuta Oshima, Masahiro Suzuki, Yutaka Matsuo
Iterative Amortized Inference Authors: ['Yuta Oshima' 'Masahiro Suzuki' 'Yutaka Matsuo'] In recent years, deep generative models for multimodal data have gained significant attention. Among these, multimodal variational autoencoders (VAEs) have emerged as a promising approach, aiming to capture a shared latent…
Alexander Tscshantz, Beren Millidge, Anil K. Seth, Christopher L. Buckley + 1 more
'Christopher L. Buckley' 'Ulrik R. Beierholm'] Predictive coding is an influential model of cortical neural activity. It proposes that perceptual beliefs are furnished by sequentially minimising “prediction errors”-the differences between predicted and observed data. Implicit in this proposal is the idea that…
Jonas Arruda, Yannik Schälte, Clemens Peiter, Olga Teplytska + 2 more
Non-linear mixed-effects models are a powerful tool for studying heterogeneous populations in various fields, including biology, medicine, economics, and engineering. Here, the aim is to find a distribution over the parameters that describe the whole population using a model that can generate simulations for an…
Lukas Schumacher, Martin Schnuerch, Andreas Voss, Stefan T. Radev
Cognitive processes undergo various fluctuations and transient states across different temporal scales. Superstatistics are emerging as a flexible framework for incorporating such non-stationary dynamics into existing cognitive model classes. In this work, we provide the first experimental validation of superstatistics…
Luca Ambrogioni, Umut Güçlü, Julia Berezutskaya, Eva W. P. van den Borne + 4 more
'Eva W. P. van den Borne' 'Yağmur Güçlütürk' 'Max Hinne' 'Eric Maris' 'Marcel van Gerven'] In this paper, we introduce a new form of amortized variational inference by using the forward KL divergence in a joint-contrastive variational loss. The resulting forward amortized variational inference is a likelihood-free…
Cole L. Hurwitz, Kai Xu, Akash Srivastava, Alessio Paolo Buccino + 1 more
Extracellular recordings using modern, dense probes provide detailed footprints of action potentials (spikes) from thousands of neurons simultaneously. Inferring the activity of single neurons from these recordings, however, is a complex blind source separation problem, complicated both by the high intrinsic data…
Zijian Wang, Jan Hasenauer, Yannik Schälte
Amortized simulation-based neural posterior estimation provides a novel machine learning based approach for solving parameter estimation problems. It has been shown to be computationally efficient and able to handle complex models and data sets. Yet, the available approach cannot handle the in experimental studies…
Zijian Wang, Jan Hasenauer, Yannik Schälte, James R. Faeder
Amortized simulation-based neural posterior estimation provides a novel machine learning based approach for solving parameter estimation problems. It has been shown to be computationally efficient and able to handle complex models and data sets. Yet, the available approach cannot handle the in experimental studies…
Tim Genewein, Jordi Grau-Moya, Li Kevin Wenliang, Laurent Orseau + 2 more
The success of large neural networks trained for sequential prediction via log-loss minimization over massive and diverse datasets has sparked debate regarding the fundamental limits of this paradigm. While these models are not explicitly programmed to perform planning and search, their behavior increasingly resembles…
Karel Veldkamp, Raoul Grasman, Dylan Molenaar
Recently Variational Autoencoders (VAEs) have been proposed as a method to estimate high dimensional Item Response Theory (IRT) models on large datasets. Although these improve the efficiency of estimation drastically compared to traditional methods, they have no natural way to deal with missing values. In this paper…
Nathan Frey, Ryan Soklaski, Simon Axelrod, Siddharth Samsi + 3 more
Massive scale, both in terms of data availability and computation, enables significant breakthroughs in key application areas of deep learning such as natural language processing (NLP) and computer vision. There is emerging evidence that scale may be a key ingredient in scientific deep learning, but the importance of…
Aryan Deshwal, Cory Simon, Janardhan Rao Doppa
Given a gas storage or separation task, we wish to search a library of nanoporous materials (NPMs) for the one with the optimal adsorption property. The high cost of measuring the adsorption property of an NPM, whether in the lab or a simulation, precludes exhaustive search. We explain, demonstrate, and advocate…
Martin Robinson, Alan Bond, Alexandr Simonov, Jie Zhang + 1 more
Recently, we have introduced the use of techniques drawn from Bayesian statistics to recover kinetic and thermodynamic parameters from voltammetric data, and were able to show that the technique of large amplitude ac voltammetry yielded significantly more accurate parameter values than the equivalent dc approach. In…