no code implementations • 1 Apr 2024 • Matthias Gerstgrasser, Rylan Schaeffer, Apratim Dey, Rafael Rafailov, Henry Sleight, John Hughes, Tomasz Korbak, Rajashree Agrawal, Dhruv Pai, Andrey Gromov, Daniel A. Roberts, Diyi Yang, David L. Donoho, Sanmi Koyejo
The proliferation of generative models, combined with pretraining on web-scale data, raises a timely question: what happens when these models are trained on their own generated outputs?
no code implementations • 15 Feb 2024 • Rylan Schaeffer, Nika Zahedi, Mikail Khona, Dhruv Pai, Sang Truong, Yilun Du, Mitchell Ostrow, Sarthak Chandra, Andres Carranza, Ila Rani Fiete, Andrey Gromov, Sanmi Koyejo
Based on the observation that associative memory's energy functions can be seen as probabilistic modeling's negative log likelihoods, we build a bridge between the two that enables useful flow of ideas in both directions.
no code implementations • 11 Jan 2024 • Minhao Jiang, Ken Ziyu Liu, Ming Zhong, Rylan Schaeffer, Siru Ouyang, Jiawei Han, Sanmi Koyejo
Language models pre-trained on web-scale corpora demonstrate impressive capabilities on diverse downstream tasks.
no code implementations • 6 Dec 2023 • Rylan Schaeffer, Mikail Khona, Sanmi Koyejo, Ila Rani Fiete
Work on deep learning-based models of grid cells suggests that grid cells generically and robustly arise from optimizing networks to path integrate, i. e., track one's spatial position by integrating self-velocity signals.
no code implementations • 5 Dec 2023 • Victor Lecomte, Kushal Thaman, Rylan Schaeffer, Naomi Bashkansky, Trevor Chow, Sanmi Koyejo
Using a combination of theory and experiments, we show that incidental polysemanticity can arise due to multiple reasons including regularization and neural noise; this incidental polysemanticity occurs because random initialization can, by chance alone, initially assign multiple features to the same neuron, and the training dynamics then strengthen such overlap.
no code implementations • 27 Nov 2023 • Rylan Schaeffer, Mikail Khona, Adrian Bertagnoli, Sanmi Koyejo, Ila Rani Fiete
At both the population and single-cell levels, we find evidence suggesting that neither of the assumptions are likely true in biological neural representations.
no code implementations • 13 Sep 2023 • Rylan Schaeffer
Inspired by recent work demonstrating the promise of smaller Transformer-based language models pretrained on carefully curated data, we supercharge such approaches by investing heavily in curating a novel, high quality, non-synthetic data mixture based solely on evaluation benchmarks.
no code implementations • 20 Jul 2023 • Dhruv Pai, Andres Carranza, Rylan Schaeffer, Arnuv Tandon, Sanmi Koyejo
We present FACADE, a novel probabilistic and geometric framework designed for unsupervised mechanistic anomaly detection in deep neural networks.
no code implementations • 20 Jul 2023 • Andres Carranza, Dhruv Pai, Rylan Schaeffer, Arnuv Tandon, Sanmi Koyejo
As the capabilities of large machine learning models continue to grow, and as the autonomy afforded to such models continues to expand, the spectre of a new adversary looms: the models themselves.
no code implementations • 20 Jul 2023 • Rylan Schaeffer, Kateryna Pistunova, Samar Khanna, Sarthak Consul, Sanmi Koyejo
We find that the logically \textit{invalid} reasoning prompts do indeed achieve similar performance gains on BBH tasks as logically valid reasoning prompts.
no code implementations • NeurIPS 2023 • Boxin Wang, Weixin Chen, Hengzhi Pei, Chulin Xie, Mintong Kang, Chenhui Zhang, Chejian Xu, Zidi Xiong, Ritik Dutta, Rylan Schaeffer, Sang T. Truong, Simran Arora, Mantas Mazeika, Dan Hendrycks, Zinan Lin, Yu Cheng, Sanmi Koyejo, Dawn Song, Bo Li
Yet, while the literature on the trustworthiness of GPT models remains limited, practitioners have proposed employing capable GPT models for sensitive applications such as healthcare and finance -- where mistakes can be costly.
no code implementations • NeurIPS 2023 • Rylan Schaeffer, Brando Miranda, Sanmi Koyejo
Recent work claims that large language models display emergent abilities, abilities not present in smaller-scale models that are present in larger-scale models.
1 code implementation • 24 Mar 2023 • Rylan Schaeffer, Mikail Khona, Zachary Robertson, Akhilan Boopathy, Kateryna Pistunova, Jason W. Rocks, Ila Rani Fiete, Oluwasanmi Koyejo
Double descent is a surprising phenomenon in machine learning, in which as the number of model parameters grows relative to the number of data, test error drops as models grow ever larger into the highly overparameterized (data undersampled) regime.
no code implementations • 2 May 2022 • Rylan Schaeffer, Gabrielle Kaili-May Liu, Yilun Du, Scott Linderman, Ila Rani Fiete
Learning from a continuous stream of non-stationary data in an unsupervised manner is arguably one of the most common and most challenging settings facing intelligent agents.
no code implementations • 5 Nov 2021 • Rylan Schaeffer
To the machine learning community, our proposed theory creates a novel interaction between the Actor and Critic in Actor-Critic agents and notes a novel connection between RL and Bayesian Optimization.
no code implementations • NeurIPS 2020 • Rylan Schaeffer, Mikail Khona, Leenoy Meshulam, Brain Laboratory International, Ila Fiete
Third, the geometry of RNN dynamics reflects an induced coupling between the two separate inference processes necessary to solve the task.