Alex Sheng

2023 From Language Models to Practical Self-Improving Agents Preprint We develop a simple and straightforward methodology to create AI computer agents that can carry out diverse computer tasks and self-improve by developing tools and augmentations to enable themselves to solve increasingly complex tasks. As large language models (LLMs) have been shown to benefit from non-parametric augmentations, a significant body of recent work has focused on developing software that augments LLMs with various capabilities. Rather than manually developing static software to augment LLMs through human engineering effort, we propose that an LLM agent can systematically generate software to augment itself. We show, through a few case studies, that a minimal querying loop with appropriate prompt engineering allows an LLM to generate and use various augmentations, freely extending its own capabilities to carry out real-world computer tasks. Starting with only terminal access, we prompt an LLM agent to augment itself with retrieval, internet search, web navigation, and text editor capabilities. The agent effectively uses these various tools to solve problems including automated software development and web-based tasks. [Link]

2023 Self-Programming Artificial Intelligence Using Code-Generating Language Models Preprint Recent progress in large-scale language models has enabled breakthroughs in previously intractable computer programming tasks. Prior work in meta-learning and neural architecture search has led to substantial successes across various task domains, spawning myriad approaches for algorithmically optimizing the design and learning dynamics of deep learning models. At the intersection of these research areas, we implement a code-generating language model with the ability to modify its own source code. Self-programming AI algorithms have been of interest since the dawn of AI itself. Although various theoretical formulations of generalized self-programming AI have been posed, no such system has been successfully implemented to date under real-world computational constraints. Applying AI-based code generation to AI itself, we develop and experimentally validate the first practical implementation of a self-programming AI system. We empirically show that a self-programming AI implemented using a code generation model can successfully modify its own source code to improve performance and program sub-models to perform auxiliary tasks. Our model can self-modify various properties including model architecture, computational capacity, and learning dynamics. [Link]

2022 Task Transfer and Domain Adaptation for Zero-Shot Question Answering NAACL Pretrained language models have shown success in various areas of natural language processing, including reading comprehension tasks. However, when applying machine learning methods to new domains, labeled data may not always be available. To address this, we use supervised pretraining on source-domain data to reduce sample complexity on domain-specific downstream tasks. We evaluate zero-shot performance on domain-specific reading comprehension tasks by combining task transfer with domain adaptation to fine-tune a pretrained model with no labelled data from the target task. Our approach outperforms Domain Adaptive Pretraining on downstream domain-specific reading comprehension tasks in 3 out of 4 domains. [Link]

2020 Distributed Evolution Strategies Using TPUs for Meta-Learning IEEE Meta-learning traditionally relies on backpropagation through entire tasks to iteratively improve a model's learning dynamics. However, this approach is computationally intractable when scaled to complex tasks. We propose a distributed evolutionary meta-learning strategy using Tensor Processing Units (TPUs) that is highly parallel and scalable to arbitrarily long tasks with no increase in memory cost. Using a Prototypical Network trained with evolution strategies on the Omniglot dataset, we achieved an accuracy of 98.4% on a 5-shot classification problem. Our algorithm used as much as 40 times less memory than automatic differentiation to compute the gradient, with the resulting model achieving accuracy within 1.3% of a backpropagation-trained equivalent (99.6%). We observed better classification accuracy as high as 99.1% with larger population configurations. We further experimentally validate the stability and performance of ES-ProtoNet across a variety of training conditions (varying population size, model size, number of workers, shot, way, ES hyperparameters, etc.). Our contributions are twofold: we provide the first assessment of evolutionary meta-learning in a supervised setting, and create a general framework for distributed evolution strategies on TPUs. [Link]

2019 Retrieval-Augmented Language Models Independent Research In high school, I developed a method to condition generative language models on external knowledge using vector search. I shared the algorithm with researchers from leading labs in industry and academia, as well as through open-source code. The widespread dissemination and popularization of the approach, including several seminal papers from various labs, spawned an active new area of research and applications. This methodology came to be known as retrieval augmentation, and is widely used today in modern language models.