Search Results for author:
Found 30 papers, 18 papers with code
tsGT: Stochastic Time Series Modeling With Transformer
Time series methods are of fundamental importance in virtually any field of science that deals with temporally structured data.
Overestimation, Overfitting, and Plasticity in Actor-Critic: the Bitter Lesson of Reinforcement Learning
We tested these agents across 14 diverse tasks from 2 simulation benchmarks.
Analysing The Impact of Sequence Composition on Language Model Pre-Training
In this work, we find that applying causal masking can lead to the inclusion of distracting information from previous documents during pre-training, which negatively impacts the performance of the models on language modelling and downstream tasks.
Fine-tuning Reinforcement Learning Models is Secretly a Forgetting Mitigation Problem
Fine-tuning is a widespread technique that allows practitioners to transfer pre-trained capabilities, as recently showcased by the successful applications of foundation models.
MoE-Mamba: Efficient Selective State Space Models with Mixture of Experts
State Space Models (SSMs) have become serious contenders in the field of sequential modeling, challenging the dominance of Transformers.
Structured Packing in LLM Training Improves Long Context Utilization
Recent developments in long-context large language models have attracted considerable attention.
Focused Transformer: Contrastive Training for Context Scaling
This novel approach enhances the structure of the (key, value) space, enabling an extension of the context length.
Exploring Continual Learning of Diffusion Models
Diffusion models have achieved remarkable success in generating high-quality images thanks to their novel training procedures applied to unprecedented amounts of data.
Magnushammer: A Transformer-Based Approach to Premise Selection
By combining \method with a language-model-based automated theorem prover, we further improve the state-of-the-art proof success rate from $57. 0\%$ to $71. 0\%$ on the PISA benchmark using $4$x fewer parameters.
The Effectiveness of World Models for Continual Reinforcement Learning
World models power some of the most efficient reinforcement learning algorithms.
Trust Your $\nabla$: Gradient-based Intervention Targeting for Causal Discovery
In this work, we propose a novel Gradient-based Intervention Targeting method, abbreviated GIT, that 'trusts' the gradient estimator of a gradient-based causal discovery framework to provide signals for the intervention acquisition function.
Disentangling Transfer in Continual Reinforcement Learning
The ability of continual learning systems to transfer knowledge from previously seen tasks in order to maximize performance on new tasks is a significant challenge for the field, limiting the applicability of continual learning solutions to realistic scenarios.
Fast and Precise: Adjusting Planning Horizon with Adaptive Subgoal Search
Complex reasoning problems contain states that vary in the computational cost required to determine a good action plan.
Thor: Wielding Hammers to Integrate Language Models and Automated Theorem Provers
Thor increases a language model's success rate on the PISA dataset from $39\%$ to $57\%$, while solving $8. 2\%$ of problems neither language models nor automated theorem provers are able to solve on their own.
Ranked #2 on
Automated Theorem Proving
on miniF2F-test
Continuous Control With Ensemble Deep Deterministic Policy Gradients
The growth of deep reinforcement learning (RL) has brought multiple exciting tools and methods to the field.
Off-Policy Correction For Multi-Agent Reinforcement Learning
The key advantage of our algorithm is its high scalability in a multi-worker setting.
Multi-agent Reinforcement Learning
reinforcement-learning
+2
Catalytic Role Of Noise And Necessity Of Inductive Biases In The Emergence Of Compositional Communication
Communication is compositional if complex signals can be represented as a combination of simpler subparts.
Subgoal Search For Complex Reasoning Tasks
In this paper, we implement kSubS using a transformer-based subgoal module coupled with the classical best-first search framework.
Continual World: A Robotic Benchmark For Continual Reinforcement Learning
Continual learning (CL) -- the ability to continuously learn, building on previously acquired knowledge -- is a natural requirement for long-lived autonomous reinforcement learning (RL) agents.
Planning and Learning Using Adaptive Entropy Tree Search
We present Adaptive Entropy Tree Search (ANTS) - a novel algorithm combining planning and learning in the maximum entropy paradigm.
Structure and randomness in planning and reinforcement learning
Planning in large state spaces inevitably needs to balance depth and breadth of the search.
Trust, but verify: model-based exploration in sparse reward environments
We propose $\textit{trust-but-verify}$ (TBV) mechanism, a new method which uses model uncertainty estimates to guide exploration.
Model-based Reinforcement Learning
reinforcement-learning
+1
CARLA Real Traffic Scenarios -- novel training ground and benchmark for autonomous driving
This work introduces interactive traffic scenarios in the CARLA simulator, which are based on real-world traffic.
Emergence of compositional language in communication through noisy channel
In this paper, we investigate how communication through a noisy channel can lead to the emergence of compositional language.
Uncertainty-sensitive Learning and Planning with Ensembles
The former manifests itself through the use of value function, while the latter is powered by a tree search planner.
Simulation-based reinforcement learning for real-world autonomous driving
Using reinforcement learning in simulation and synthetic data is motivated by lowering costs and engineering effort.
Developmentally motivated emergence of compositional communication via template transfer
This paper explores a novel approach to achieving emergent compositional communication in multi-agent systems.
Uncertainty - sensitive learning and planning with ensembles
Notably, our method performs well in environments with sparse rewards where standard $TD(1)$ backups fail.
Expert-augmented actor-critic for ViZDoom and Montezumas Revenge
We propose an expert-augmented actor-critic algorithm, which we evaluate on two environments with sparse rewards: Montezumas Revenge and a demanding maze from the ViZDoom suite.
Learning to Run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments
In the NIPS 2017 Learning to Run challenge, participants were tasked with building a controller for a musculoskeletal model to make it run as fast as possible through an obstacle course.
