Arash Tavakoli

5 papers · 2018–2022 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (12) 🧭 Keyword Pioneer

Venues

ICLR (3) ICML (1) NeurIPS (1)

Top co-authors

Mehdi Fatemi (3) Petar Kormushev (2) Vitaly Levdik (1) Dimitrije Antić (1) Fabio Pardo (1) Georg Martius (1) Harm Van Seijen (1) Maximilian Seitzer (1)

Research topics

Machine Learning (4) Reinforcement Learning (4) Methods (4) Deep Learning (3) Artificial Intelligence (2) Applications (2) Learning Types · Machine Learning (2) Optimization & Theory (2) Architectures (1) Bayesian & Probabilistic (1) Core AI (1) Learning Types · Deep Learning (1) Models (1)

Keywords

policy optimization (2) reinforcement learning (2) value function (2) discount factor (1) hypergraphs (1) maximum likelihood estimation (1) variance estimation (1) value estimation (1) gaussian distribution (1) markov property (1) control task (1) bayesian neural network (1) experience replay (1) reward decomposition (1) time limit (1) episodic training (1) combinatorial structure (1) convergence proof (1) heteroscedastic uncertainty (1) action gap (1)

Papers

On the Pitfalls of Heteroscedastic Uncertainty Estimation with Probabilistic Neural Networks ICLR 2022 Orchestrated Value Mapping for Reinforcement Learning ICLR 2022 Learning to Represent Action Values as a Hypergraph on the Action Vertices ICLR 2021 Using a Logarithmic Mapping to Enable Lower Discount Factors in Reinforcement Learning NeurIPS 2019 Time Limits in Reinforcement Learning ICML 2018