The Malleable Agent

Reliable Communication Through a Noisy Channel

The limits of my prompt

Tokens for nothing and your (cognitive) chips for free

Joblib for memoization

GitHub actions 101

Conformal Classification of Neural Networks

Back to Square One: Artifact Detection, Training and Commonsense Disentanglement in the Winograd Schema

Probing Emergent Semantics in Predictive Agents via Question Answering

KG-based world models

Learning salon: Kimberly Stachenfeld

Learning salon: Jane Wang

Simple Few-Shot Learning with Language Models

Learning to communicate about shared procedural abstractions

Learning Salon: Recap & Discussion

Reward is enough

Making Sense of Sensory Input; Apperception code

Learning salon: Zenna Tavares

Ideas about a new programming language for games

Are Convolutional NNs or Transformers more like Human Vision?

Flexible modulation of sequence generation in the entorhinal–hippocampal system

Studying and prompts

Python servers and WSGI

Storing episodic memory: place and grid cells

Building Athens Research (Clojure/JS)

Emacs config

Learning Salon: Simon Kornblith

MuZero

Kotlin codebase

Learning Salon: Blake Richards

Semantic Grounding of Novel Spoken Words in the Primary Visual Cortex

Learning Salon: Katja Hofmann, Deep RL for Games

BERT objectives

Learning Salon: Recap and Panel Discussion

Making Sense of Sensory Input

Symbolic Behavior in AI

Predicting Transferability in NLP

Object-based attention for spatio-temporal reasoning

Go Explore

Autonomous navigation of stratospheric looney tunes using RL

Levels of biological plausibility

If Deep Learning is the answer, what is the question?

Psi: A Computational Architecture

Learning to Speak and Act in a Fantasy Text Adventure Game

Importance-weighted Actor-Learner Architecture

Placing language in an integrated understanding system: Next steps toward human-level performance in neural language models

Frame semantics

Compositional Reasoning in Early Childhood

An agenda for disinformation research

[AI4Good 2020] Don't ExplAIn yourself

Environmental drivers of systematicity and generalisation in a situated agent

Language Wars

Algorithms for Causal Reasoning in Probability Trees

To the Jupyter config and back

Towards Embodied Language Understanding

Emergent Communications Pretraining for Few-Shot NMT

Dat Python, dataclass

Lists of NumPy arrays

Money for nothing and ad-hoc objects for free

Catching a main class

Lost in Python

The dark ages of `C++20` string formatting