Publications

Language Model Alignment with Elastic Reset
Michael Noukhovitch
Samuel Lavoie
Florian Strub
Finetuning language models with reinforcement learning (RL), e.g. from human feedback (HF), is a prominent method for alignment. But optimiz… (voir plus)ing against a reward model can improve on reward while degrading performance in other areas, a phenomenon known as reward hacking, alignment tax, or language drift. First, we argue that commonly-used test metrics are insufficient and instead measure how different algorithms tradeoff between reward and drift. The standard method modified the reward with a Kullback-Lieber (KL) penalty between the online and initial model. We propose Elastic Reset, a new algorithm that achieves higher reward with less drift without explicitly modifying the training objective. We periodically reset the online model to an exponentially moving average (EMA) of itself, then reset the EMA model to the initial model. Through the use of an EMA, our model recovers quickly after resets and achieves higher reward with less drift in the same number of steps. We demonstrate that fine-tuning language models with Elastic Reset leads to state-of-the-art performance on a small scale pivot-translation benchmark, outperforms all baselines in a medium-scale RLHF-like IMDB mock sentiment task and leads to a more performant and more aligned technical QA chatbot with LLaMA-7B. Code available at github.com/mnoukhov/elastic-reset.
Laughing Hyena Distillery: Extracting Compact Recurrences From Convolutions
Stefano Massaroli
Michael Poli
Daniel Y Fu
Hermann Kumbong
Rom Nishijima Parnichkun
Aman Timalsina
David W. Romero
Quinn McIntyre
Beidi Chen
Atri Rudra
Ce Zhang
Christopher Re
Stefano Ermon
Recent advances in attention-free sequence models rely on convolutions as alternatives to the attention operator at the core of Transformers… (voir plus). In particular, long convolution sequence models have achieved state-of-the-art performance in many domains, but incur a significant cost during auto-regressive inference workloads -- naively requiring a full pass (or caching of activations) over the input sequence for each generated token -- similarly to attention-based models. In this paper, we seek to enable
Learning better with Dale’s Law: A Spectral Perspective
Pingsheng Li
Jonathan Cornford
Arna Ghosh
Learning Reliable Logical Rules with SATNet
Zhaoyu Li
Jinpei Guo
Yuhe Jiang
Let the Flows Tell: Solving Graph Combinatorial Problems with GFlowNets
Dinghuai Zhang
Hanjun Dai
Nikolay Malkin
Ling Pan
Lie Point Symmetry and Physics-Informed Networks
Tara Akhound-Sadegh
Johannes Brandstetter
Max Welling
Symmetries have been leveraged to improve the generalization of neural networks through different mechanisms from data augmentation to equiv… (voir plus)ariant architectures. However, despite their potential, their integration into neural solvers for partial differential equations (PDEs) remains largely unexplored. We explore the integration of PDE symmetries, known as Lie point symmetries, in a major family of neural solvers known as physics-informed neural networks (PINNs). We propose a loss function that informs the network about Lie point symmetries in the same way that PINN models try to enforce the underlying PDE through a loss function. Intuitively, our symmetry loss ensures that the infinitesimal generators of the Lie group conserve the PDE solutions.. Effectively, this means that once the network learns a solution, it also learns the neighbouring solutions generated by Lie point symmetries. Empirical evaluations indicate that the inductive bias introduced by the Lie point symmetries of the PDEs greatly boosts the sample efficiency of PINNs.
Maximum State Entropy Exploration using Predecessor and Successor Representations
Arnav Kumar Jain
Lucas Lehnert
Animals have a developed ability to explore that aids them in important tasks such as locating food, exploring for shelter, and finding misp… (voir plus)laced items. These exploration skills necessarily track where they have been so that they can plan for finding items with relative efficiency. Contemporary exploration algorithms often learn a less efficient exploration strategy because they either condition only on the current state or simply rely on making random open-loop exploratory moves. In this work, we propose
Multi-Head Adapter Routing for Cross-Task Generalization
Lucas Caccia
Edoardo Ponti
Zhan Su
Matheus Pereira
Parameter-efficient fine-tuning (PEFT) for cross-task generalization consists in pre-training adapters on a multi-task training set before f… (voir plus)ew-shot adaptation to test tasks. Polytropon [Ponti et al., 2023] (
Neural Graph Generation from Graph Statistics
Kiarash Zahirnia
Yaochen Hu
Oliver Schulte
Neural Graph Generation from Graph Statistics.
Kiarash Zahirnia
Yaochen Hu
Oliver Schulte
Optimal Extragradient-Based Algorithms for Stochastic Variational Inequalities with Separable Structure
Angela Yuan
Chris Junchi Li
Michael Jordan
Quanquan Gu
Simon Shaolei Du
We consider the problem of solving stochastic monotone variational inequalities with a separable structure using a stochastic first-order or… (voir plus)acle. Building on standard extragradient for variational inequalities we propose a novel algorithm---stochastic \emph{accelerated gradient-extragradient} (AG-EG)---for strongly monotone variational inequalities (VIs). Our approach combines the strengths of extragradient and Nesterov acceleration. By showing that its iterates remain in a bounded domain and applying scheduled restarting, we prove that AG-EG has an optimal convergence rate for strongly monotone VIs. Furthermore, when specializing to the particular case of bilinearly coupled strongly-convex-strongly-concave saddle-point problems, including bilinear games, our algorithm achieves fine-grained convergence rates that match the respective lower bounds, with the stochasticity being characterized by an additive statistical error term that is optimal up to a constant prefactor.
Parallel-mentoring for Offline Model-based Optimization
Can Chen
Christopher Beckham
Zixuan Liu
We study offline model-based optimization to maximize a black-box objective function with a static dataset of designs and scores. These desi… (voir plus)gns encompass a variety of domains, including materials, robots, DNA sequences, and proteins. A common approach trains a proxy on the static dataset and performs gradient ascent to obtain new designs. However, this often results in poor designs due to the proxy inaccuracies for out-of-distribution designs. Recent studies indicate that (a) gradient ascent with a mean ensemble of proxies generally outperforms simple gradient ascent, and (b) a trained proxy provides weak ranking supervision signals for design selection. Motivated by (a) and (b), we propose