59 citations · 134 across the 21 of their papers we have counts for
18 papers · 1 filter
Training AI Scientists to Replicate Research
Damon Falck, Samer Sabri, Anja Surina +8
The replicability of papers is a cornerstone of scientific knowledge, ensuring the reliability of existing results and providing a base for further experiments. The act of replicat…
Understanding In-Context Learning of Linear Models in Transformers Through an Adversarial Lens
Usman Anwar, Johannes Von Oswald, Louis Kirsch +2
In this work, we make two contributions towards understanding of in-context learning of linear models by transformers. First, we investigate the adversarial robustness of in-contex…
Sequence Compression Speeds Up Credit Assignment in Reinforcement Learning
Aditya A. Ramesh, Kenny Young, Louis Kirsch +1
Temporal credit assignment in reinforcement learning is challenging due to delayed and stochastic outcomes. Monte Carlo targets can bridge long delays between action and consequenc…
Discovering Temporally-Aware Reinforcement Learning Algorithms
Matthew Thomas Jackson, Chris Lu, Louis Kirsch +3
Recent advancements in meta-learning have enabled the automatic discovery of novel reinforcement learning algorithms parameterized by surrogate objective functions. To improve upon…
The Languini Kitchen: Enabling Language Modelling Research at Different Scales of Compute
Aleksandar Stanić, Dylan Ashley, Oleg Serikov +5
The Languini Kitchen serves as both a research collective and codebase designed to empower researchers with limited computational resources to contribute meaningfully to the field…
Eliminating Meta Optimization Through Self-Referential Meta Learning
Louis Kirsch, Jürgen Schmidhuber
Meta Learning automates the search for learning algorithms. At the same time, it creates a dependency on human engineering on the meta-level, where meta learning algorithms need to…
Learning One Abstract Bit at a Time Through Self-Invented Experiments Encoded as Neural Networks
Vincent Herrmann, Louis Kirsch, Jürgen Schmidhuber
There are two important things in science: (A) Finding answers to given questions, and (B) Coming up with good questions. Our artificial scientists not only learn to answer given q…
General-Purpose In-Context Learning by Meta-Learning Transformers
Louis Kirsch, James Harrison, Jascha Sohl-Dickstein +1
Modern machine learning requires system designers to specify aspects of the learning pipeline, such as losses, architectures, and optimizers. Meta-learning, or learning-to-learn, i…
Exploring through Random Curiosity with General Value Functions
Aditya Ramesh, Louis Kirsch, Sjoerd van Steenkiste +1
Efficient exploration in reinforcement learning is a challenging problem commonly addressed through intrinsic rewards. Recent prominent approaches are based on state novelty or var…
The Benefits of Model-Based Generalization in Reinforcement Learning
Kenny Young, Aditya Ramesh, Louis Kirsch +1
Model-Based Reinforcement Learning (RL) is widely believed to have the potential to improve sample efficiency by allowing an agent to synthesize large amounts of imagined experienc…
Goal-Conditioned Generators of Deep Policies
Francesco Faccio, Vincent Herrmann, Aditya Ramesh +2
Goal-conditioned Reinforcement Learning (RL) aims at learning optimal policies, given goals encoded in special command inputs. Here we study goal-conditioned neural nets (NNs) that…
Introducing Symmetries to Black Box Meta Reinforcement Learning
Louis Kirsch, Sebastian Flennerhag, Hado van Hasselt +3
Meta reinforcement learning (RL) attempts to discover new RL algorithms automatically from environment interaction. In so-called black-box approaches, the policy and the learning a…
Meta Learning Backpropagation And Improving It
Louis Kirsch, Jürgen Schmidhuber
Many concepts have been proposed for meta learning with neural networks (NNs), e.g., NNs that learn to reprogram fast weights, Hebbian plasticity, learned learning rules, and meta…
Parameter-Based Value Functions
Francesco Faccio, Louis Kirsch, Jürgen Schmidhuber
Traditional off-policy actor-critic Reinforcement Learning (RL) algorithms learn value functions of a single target policy. However, when value functions are updated to track the l…
Improving Generalization in Meta Reinforcement Learning using Learned Objectives
Louis Kirsch, Sjoerd van Steenkiste, Jürgen Schmidhuber
Biological evolution has distilled the experiences of many learners into the general learning algorithms of humans. Our novel meta reinforcement learning algorithm MetaGenRL is ins…
Gaussian Mean Field Regularizes by Limiting Learned Information
Julius Kunze, Louis Kirsch, Hippolyt Ritter +1
Variational inference with a factorized Gaussian posterior estimate is a widely used approach for learning parameters and hidden variables. Empirically, a regularizing effect can b…
Modular Networks: Learning to Decompose Neural Computation
Louis Kirsch, Julius Kunze, David Barber
Scaling model capacity has been vital in the success of deep learning. For a typical network, necessary compute resources and training time grow dramatically with model size. Condi…
Transfer Learning for Speech Recognition on a Budget
Julius Kunze, Louis Kirsch, Ilia Kurenkov +3
End-to-end training of automated speech recognition (ASR) systems requires massive data and compute resources. We explore transfer learning based on model adaptation as an approach…