WES: Agent-based User Interaction Simulation on Real Infrastructure

Ahlgren, John; Berezin, Maria Eugenia; Bojarczuk, Kinga; Dulskyte, Elena; Dvortsova, Inna; George, Johann; Gucevska, Natalija; Harman, Mark; Lämmel, Ralf; Meijer, Erik; Sapora, Silvia; Spahr-Summers, Justin

doi:10.48550/arxiv.2004.05363

Cited by 4 publications

(9 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generative adversarial networks have been used to generate virtual users for high-fidelity recommender environments to support learning policies that can be transferred to real systems [44,62]. Multi-agent simulation has a long history, and has recently found use for modeling agent interactions in economic [64], social media [1] and social network settings [8,53]. None of these models support the authoring of general interaction models.…”

Section: Related Workmentioning

confidence: 99%

“…The RecSim NG library is available for download, 1 and more detailed exposition of it capabilities and its usage can be found in the tutorials offered there. Some of the details of the examples and use cases presented in the sequel are left unstated, but full details (including models, algorithms and their specific parametrizations) can be found in the library.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Demonstrating Principled Uncertainty Modeling for Recommender Ecosystems with RecSim NG

Mladenov

Hsu

Jain

et al. 2020

Fourteenth ACM Conference on Recommender Systems

View full text Add to dashboard Cite

The development of recommender systems that optimize multi-turn interaction with users, and model the interactions of different agents (e.g., users, content providers, vendors) in the recommender ecosystem have drawn increasing attention in recent years.Developing and training models and algorithms for such recommenders can be especially difficult using static datasets, which often fail to offer the types of counterfactual predictions needed to evaluate policies over extended horizons. To address this, we develop RecSim NG, a probabilistic platform for the simulation of multi-agent recommender systems. RecSim NG is a scalable, modular, differentiable simulator implemented in Edward2 and TensorFlow. It offers: a powerful, general probabilistic programming language for agent-behavior specification; tools for probabilistic inference and latent-variable model learning, backed by automatic differentiation and tracing; and a TensorFlow-based runtime for running simulations on accelerated hardware. We describe RecSim NG and illustrate how it can be used to create transparent, configurable, end-to-end models of a recommender ecosystem, complemented by a small set of simple use cases that demonstrate how RecSim NG can help both researchers and practitioners easily develop and train novel algorithms for recommender systems.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Demonstrating Principled Uncertainty Modeling for Recommender Ecosystems with RecSim NG

Mladenov

Hsu

Jain

et al. 2020

Fourteenth ACM Conference on Recommender Systems

View full text Add to dashboard Cite

show abstract

“…We hope that this paper re-starts the discussion on a technical level, with ever increasing planning methods and simulations coming to light. There is no reason for the plan to be as simple as the one discussed here -in fact Facebook is currently performing large scale simulations [Ahlgren et al, 2020]; one can easily envision a situation where simulated production/consumption unit behaviour is used to plan in a setting more closely aligned to traditional RL.…”

Section: Discussionmentioning

confidence: 99%

Open Loop In Natura Economic Planning

Samothrakis

2020

Preprint

View full text Add to dashboard Cite

The debate between the optimal way of allocating societal surplus (i.e. products and services) has been raging, in one form or another, practically forever; following the collapse of the Soviet Union in 1991, the market became the only legitimate form of organisation -there was no other alternative. Working within the tradition of Marx, Leontief, Kantorovich, Beer and Cockshott, we propose what we deem an automated planning system that aims to operate on unit level (e.g., factories and citizens), rather than on aggregate demand and sectors. We explain why it is both a viable and desirable alternative to current market conditions and position our solution within current societal structures. Our experiments show that it would be trivial to plan for up to 50K industrial goods and 5K final goods in commodity hardware.1 Some readers might complain that we conflate markets with capitalism. We are doing this on purpose. 2 We are not really that naive. Power politics abide.Preprint. Under review.

show abstract

“…In adjacent technologies, simulations of individual behaviour on social media platforms are being used to predict reactions to changes in the infrastructure of the platform and to optimise the platform to nudge or elicit particular behaviours (Ahlgren et al, 2020).…”

Section: Examplesmentioning

confidence: 99%

Ethical and social risks of harm from Language Models

Weidinger¹,

Mellor²,

Rauh³

et al. 2021

Preprint

105

View full text Add to dashboard Cite

This paper aims to help structure the risk landscape associated with large-scale Language Models (LMs). In order to foster advances in responsible innovation, an in-depth understanding of the potential risks posed by these models is needed. A wide range of established and anticipated risks are analysed in detail, drawing on multidisciplinary literature from computer science, linguistics, and social sciences.

show abstract

WES: Agent-based User Interaction Simulation on Real Infrastructure

Cited by 4 publications

References 0 publications

Demonstrating Principled Uncertainty Modeling for Recommender Ecosystems with RecSim NG

Demonstrating Principled Uncertainty Modeling for Recommender Ecosystems with RecSim NG

Open Loop In Natura Economic Planning

Ethical and social risks of harm from Language Models

Contact Info

Product

Resources

About