Constructing AI models that respond to text instructions is challenging,...
Acquiring high-quality data for training discriminative models is a cruc...
Large Language Models (LLMs) present immense potential in the medical fi...
Prompt tuning is one of the successful approaches for parameter-efficien...
We propose TR0N, a highly general framework to turn pre-trained uncondit...
Methods such as chain-of-thought prompting and self-consistency have pus...
General intelligence requires solving tasks across many domains. Current...
Variational autoencoders (VAEs) are powerful tools for learning latent
r...
By conditioning on natural language instructions, large language models
...
Training deep neural networks in low rank, i.e. with factorised layers, ...
Dataset distillation aims to learn a small synthetic dataset that preser...
Recently, methods such as Decision Transformer that reduce reinforcement...
Deep Reinforcement Learning (RL) is successful in solving many complex M...
Deep learning has enabled algorithms to generate realistic images. Howev...
While designing inductive bias in neural architectures has been widely
s...
The current success of deep learning depends on large-scale labeled data...
Reinforcement learning has enabled agents to solve challenging tasks in
...
Learning task-agnostic dynamics models in high-dimensional observation s...
Intelligent agents need to generalize from past experience to achieve go...
We introduce a unified objective for action and perception of intelligen...
The impact of gradient noise on training deep models is widely acknowled...
In this work, we focus on an analogical reasoning task that contains ric...
In learning-assisted theorem proving, one of the most critical challenge...
What goals should a multi-goal reinforcement learning agent pursue durin...
While second order optimizers such as natural gradient descent (NGD) oft...
Ensembles, where multiple neural networks are trained individually and t...
Distances are pervasive in machine learning. They serve as similarity
me...
Learned world models summarize an agent's experience to facilitate learn...
Many tasks in modern machine learning can be formulated as finding equil...
The vast majority of successful deep neural networks are trained using
v...
Model-based reinforcement learning (MBRL) is widely seen as having the
p...
Model-based reinforcement learning (MBRL) with model-predictive control ...
Despite the recent successes in robotic locomotion control, the design o...
We introduce graph normalizing flows: a new, reversible graph neural net...
The choice of batch-size in a stochastic optimization algorithm plays a
...
Building agents to interact with the web would allow for significant
imp...
Sparse reward is one of the most challenging problems in reinforcement
l...
Recurrent neural networks (RNNs) provide state-of-the-art performance in...
Stochastic neural net weights are used in a variety of contexts, includi...
Generative Adversarial Networks (GANs) are one of the most practical
str...
In this work, we propose to apply trust region optimization to deep
rein...
Until recently, research on artificial neural networks was largely restr...
Despite their success, convolutional neural networks are computationally...
One of the main challenges in Zero-Shot Learning of visual categories is...
Inspired by recent work in machine translation and object detection, we
...
We present an attention-based model for recognizing multiple objects in
...