[ICML 2020] PyTorch Code for "One Policy to Control Them All: Shared Modular Policies for Agent-Agnostic Control"

[ICLR 2022] Linking Emergent and Natural Languages via Corpus Transfer

Official PyTorch (Lightning) implementation of the NeurIPS 2020 paper "Efficient Marginalization of Discrete and Structured Latent Variables via Sparsity".

PyTorch implementation of "PatchGame: Learning to Signal Mid-level Patches in Referential Games" to appear in NeurIPS 2021

RL environment replicating the werewolf game to study emergent communication

Code repository for On the interaction between supervision and self-play in emergent communication (ICLR 2020)

Emergent Communication Pretraining for Few-Shot Machine Translation

The code repository for our paper "Shared Spatial Memory Through Predictive Coding".

Code for Emergent Communication under Competition (AAMAS 2021)

Studying emergent communication among agents in cooperative and competitive environments using Reinforcement Learning, game theory and replicator dynamics

Code accompanying the paper Measuring non-trivial compositionality in emergent communication

Code repository for Capacity, Bandwidth, and Compositionality in Emergent Language Learning (AAMAS 2020)

Emergent Communication in RLlib

Source code for our paper "Learning to Communicate with Strangers via Channel Randomisation Methods"

Code accompanying the blog post "Introduction to Lewis signaling games with Python"

Project describing how social complexity relates to communication complexity

Emergent communication in multi-agent RL

Submission for EVOLANG 2026 "An Emergent Communication Framework for Honeybee Waggle Dance"

Implementation of the Asynchronous Advantage Actor Critic with Communication in TensorFlow 2

An offline-first, tactical survival terminal and emergency mesh communication system built for Android using Jetpack Compose.

Research project seeking to bridge the gap between representation learning and emergent communication

A preliminary study of emergent molecular communication protocols learned by graph-based agents in a diffusion channel environment.

Code acompanying the paper Developmentally motivated emergence of compositional communication via template transfer

Problem suite focused on evaluating the ability of agents to develop emergent communication in cooperative navigation tasks

Experimental harness for the Δ-Variable Theory of Interrogative Emergence — multi-agent MARL system with Protocol registry, Lab Notebook, and Neural Loom visualizer.

A study of convergence of agents to a communal emergent protocol. Submitted as a part of the dissertation requirements for an MSc in Cognitive Science (2024)

No description provided.