Publications

GOAt: Explaining Graph Neural Networks via Graph Output Attribution
Shengyao Lu
Keith G Mills
Jiao He
Bang Liu
Di Niu
Understanding the decision-making process of Graph Neural Networks (GNNs) is crucial to their interpretability. Most existing methods for ex… (see more)plaining GNNs typically rely on training auxiliary models, resulting in the explanations remain black-boxed. This paper introduces Graph Output Attribution (GOAt), a novel method to attribute graph outputs to input graph features, creating GNN explanations that are faithful, discriminative, as well as stable across similar samples. By expanding the GNN as a sum of scalar products involving node features, edge features and activation patterns, we propose an efficient analytical method to compute contribution of each node or edge feature to each scalar product and aggregate the contributions from all scalar products in the expansion form to derive the importance of each node and edge. Through extensive experiments on synthetic and real-world data, we show that our method not only outperforms various state-of-the-art GNN explainers in terms of the commonly used fidelity metric, but also exhibits stronger discriminability, and stability by a remarkable margin.
2024-01-15
ICLR.cc/2024/Conference (poster)
openreview.net
openreview.net
GrowSpace: A reinforcement learning environment for plant architecture
Yasmeen Hitti
Ionelia Buzatu
Manuel Del Verme
Mark Lefsrud
Florian Golemo
Audrey Durand
2024-01-15
Computers and Electronics in Agriculture (published)
doi.org
Hallucination Detection and Hallucination Mitigation: An Investigation
Junliang Luo
Tianyu Li
Di Wu
M. Jenkin
Steve Liu
Gregory Dudek
2024-01-15
ArXiv (preprint)
doi.org
arxiv.org
Improving Intrinsic Exploration by Creating Stationary Objectives
Roger Creus Castanyer
Joshua Romoff
Glen Berseth
2024-01-15
ICLR.cc/2024/Conference (poster)
doi.org
openreview.net
INViTE: INterpret and Control Vision-Language Models with Text Explanations
Haozhe Chen
Junfeng Yang
Carl Vondrick
Chengzhi Mao
Columbia University
M. University
Large-scale pre-trained vision foundation models, such as CLIP, have become de facto backbones for various vision tasks. However, due to the… (see more)ir black-box nature, understanding the underlying rules behind these models’ predictions and controlling model behaviors have remained open challenges. We present INViTE: a framework for INterpreting Vision Transformer’s latent tokens with Text Explanations. Given a latent token, INViTE retains its semantic information to the final layer using transformer’s local operations and retrieves the closest text for explanation. INViTE enables understanding of model visual reasoning procedure without needing additional model training or data collection. Based on the obtained interpretations, INViTE allows for model editing that controls model reasoning behaviors and improves model robustness against biases and spurious correlations. Our code is available at https://github.com/tonychenxyz/vit-interpret.
2024-01-15
ICLR.cc/2024/Conference (poster)
openreview.net
openreview.net
Jointly-Learned Exit and Inference for a Dynamic Neural Network
Florence Regol
Joud Chataoui
Mark J. Coates
2024-01-15
ICLR.cc/2024/Conference (poster)
openreview.net
openreview.net
Large Language Models as Generalizable Policies for Embodied Tasks
Andrew Szot
Max Schwarzer
Harsh Agrawal
Bogdan Mazoure
Walter Talbott
Rin Metcalf
Natalie Mackraz
R Devon Hjelm
Alexander T Toshev
2024-01-15
ICLR.cc/2024/Conference (poster)
doi.org
openreview.net
Mastering Memory Tasks with World Models
Mohammad Reza Samsami
Artem Zholus
Janarthanan Rajendran
A. Chandar
Current model-based reinforcement learning (MBRL) agents struggle with long-term dependencies. This limits their ability to effectively solv… (see more)e tasks involving extended time gaps between actions and outcomes, or tasks demanding the recalling of distant observations to inform current actions. To improve temporal coherence, we integrate a new family of state space models (SSMs) in world models of MBRL agents to present a new method, Recall to Imagine (R2I). This integration aims to enhance both long-term memory and long-horizon credit assignment. Through a diverse set of illustrative tasks, we systematically demonstrate that R2I not only establishes a new state-of-the-art for challenging memory and credit assignment RL tasks, such as BSuite and POPGym, but also showcases superhuman performance in the complex memory domain of Memory Maze. At the same time, it upholds comparable performance in classic RL tasks, such as Atari and DMC, suggesting the generality of our method. We also show that R2I is faster than the state-of-the-art MBRL method, DreamerV3, resulting in faster wall-time convergence.
2024-01-15
ICLR.cc/2024/Conference (oral)
doi.org
openreview.net
Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks
Samyak Jain
Robert Kirk
Ekdeep Singh Lubana
Robert P. Dick
Hidenori Tanaka
Tim Rocktäschel
Edward Grefenstette
David M. Krueger
Fine-tuning large pre-trained models has become the de facto strategy for developing both task-specific and general-purpose machine learning… (see more) systems, including developing models that are safe to deploy. Despite its clear importance, there has been minimal work that explains how fine-tuning alters the underlying capabilities learned by a model during pretraining: does fine-tuning yield entirely novel capabilities or does it just modulate existing ones? We address this question empirically in synthetic, controlled settings where we can use mechanistic interpretability tools (e.g., network pruning and probing) to understand how the model's underlying capabilities are changing. We perform an extensive analysis of the effects of fine-tuning in these settings, and show that: (i) fine-tuning rarely alters the underlying model capabilities; (ii) a minimal transformation, which we call a `wrapper', is typically learned on top of the underlying model capabilities, creating the illusion that they have been modified; and (iii) further fine-tuning on a task where such ``wrapped capabilities'' are relevant leads to sample-efficient revival of the capability, i.e., the model begins reusing these capabilities after only a few gradient steps. This indicates that practitioners can unintentionally remove a model's safety wrapper merely by fine-tuning it on a, e.g., superficially unrelated, downstream task. We additionally perform analysis on language models trained on the TinyStories dataset to support our claims in a more realistic setup.
2024-01-15
ICLR.cc/2024/Conference (poster)
openreview.net
openreview.net
Poly-View Contrastive Learning
Amitis Shidani
R Devon Hjelm
Jason Ramapuram
Russell Webb
Eeshan Gunesh Dhekane
Dan Busbridge
2024-01-15
ICLR.cc/2024/Conference (poster)
openreview.net
openreview.net
Raidar: geneRative AI Detection viA Rewriting
Chengzhi Mao
Carl Vondrick
Hao Wang
Junfeng Yang
We find that large language models (LLMs) are more likely to modify human-written text than AI-generated text when tasked with rewriting. Th… (see more)is tendency arises because LLMs often perceive AI-generated text as high-quality, leading to fewer modifications. We introduce a method to detect AI-generated content by prompting LLMs to rewrite text and calculating the editing distance of the output. We dubbed our geneRative AI Detection viA Rewriting method Raidar. Raidar significantly improves the F1 detection scores of existing AI content detection models -- both academic and commercial -- across various domains, including News, creative writing, student essays, code, Yelp reviews, and arXiv papers, with gains of up to 29 points. Operating solely on word symbols without high-dimensional features, our method is compatible with black box LLMs, and is inherently robust on new content. Our results illustrate the unique imprint of machine-generated text through the lens of the machines themselves.
2024-01-15
ICLR.cc/2024/Conference (poster)
doi.org
openreview.net
Reward Model Ensembles Help Mitigate Overoptimization
Thomas Coste
Usman Anwar
Robert Kirk
David M. Krueger
Reinforcement learning from human feedback (RLHF) is a standard approach for fine-tuning large language models to follow instructions. As pa… (see more)rt of this process, learned reward models are used to approximately model human preferences. However, as imperfect representations of the “true” reward, these learned reward models are susceptible to overoptimization. Gao et al. (2023) studied this phenomenon in a synthetic human feedback setup with a significantly larger “gold” reward model acting as the true reward (instead of humans) and showed that overoptimization remains a persistent problem regardless of the size of the proxy reward model and training data used. Using a similar setup, we conduct a systematic study to evaluate the efficacy of using ensemble-based conservative optimization objectives, specifically worst-case optimization (WCO) and uncertainty-weighted optimization (UWO), for mitigating reward model overoptimization when using two optimization methods: (a) best-of-n sampling (BoN) (b) proximal policy optimization (PPO). We additionally extend the setup of Gao et al. (2023) to include 25% label noise to better mirror real-world conditions. Both with and without label noise we find that conservative optimization practically eliminates overoptimization and improves performance by up to 70% for BoN sampling. For PPO, ensemble-based conservative optimization always reduces overoptimization and outperforms single reward model optimization. Moreover, combining it with a small KL penalty successfully prevents overoptimization at no performance cost. Overall, our results demonstrate that ensemble-based conservative optimization can effectively counter overoptimization.
2024-01-15
ICLR.cc/2024/Conference (poster)
openreview.net
openreview.net