Portrait de Michael Noukhovitch n'est pas disponible

Michael Noukhovitch

noukhovm@mila.quebec
Doctorat - Université de Montréal
Superviseur⋅e principal⋅e
Aaron Courville
Site web
Github
Google Scholar

Publications

Language Model Alignment with Elastic Reset
Michael Noukhovitch
Samuel Lavoie
Florian Strub
Aaron Courville
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.
openreview.net
Simplicial Embeddings in Self-Supervised Learning and Downstream Classification
Samuel Lavoie
Christos Tsirigotis
Max Schwarzer
Kenji Kawaguchi
Ankit Vani
Michael Noukhovitch
Aaron Courville
Simplicial Embeddings (SEM) are representations learned through self-supervised learning (SSL), wherein a representation is projected into …
2023-02-01
ICLR.cc/2023/Conference (notable)
doi.org
openreview.net
DATA-EFFICIENT REINFORCEMENT LEARNING
Nitarshan Rajkumar
Michael Noukhovitch
Ankesh Anand
Laurent Charlin
(Rex) Devon Hjelm
Philip Bachman
Aaron Courville
Data efficiency poses a major challenge for deep reinforcement learning. We approach this issue from the perspective of self-supervised repr… (voir plus)esentation learning, leveraging reward-free exploratory data to pretrain encoder networks. We employ a novel combination of latent dynamics modelling and goal-reaching objectives, which exploit the inherent structure of data in reinforcement learning. We demonstrate that our method scales well with network capacity and pretraining data. When evaluated on the Atari 100k data-efficiency benchmark, our approach significantly outperforms previous methods combining unsupervised pretraining with task-specific finetuning, and approaches human-level performance.
2021-01-01
(publié)
www.semanticscholar.org
Pretraining Representations for Data-Efficient Reinforcement Learning
Max Schwarzer
Nitarshan Rajkumar
Michael Noukhovitch
Ankesh Anand
Laurent Charlin
(Rex) Devon Hjelm
Philip Bachman
Aaron Courville
Data efficiency is a key challenge for deep reinforcement learning. We address this problem by using unlabeled data to pretrain an encoder w… (voir plus)hich is then finetuned on a small amount of task-specific data. To encourage learning representations which capture diverse aspects of the underlying MDP, we employ a combination of latent dynamics modelling and unsupervised goal-conditioned RL. When limited to 100k steps of interaction on Atari games (equivalent to two hours of human experience), our approach significantly surpasses prior work combining offline representation pretraining with task-specific finetuning, and compares favourably with other pretraining methods that require orders of magnitude more data. Our approach shows particular promise when combined with larger models as well as more diverse, task-aligned observational data -- approaching human-level performance and data-efficiency on Atari in our best setting.
openreview.net
Commonsense mining as knowledge base completion? A study on the impact of novelty
Stanisław Jastrzębski
Dzmitry Bahdanau
Seyedarian Hosseini
Michael Noukhovitch
Yoshua Bengio
Jackie Cheung
Commonsense knowledge bases such as ConceptNet represent knowledge in the form of relational triples. Inspired by recent work by Li et al., … (voir plus)we analyse if knowledge base completion models can be used to mine commonsense knowledge from raw text. We propose novelty of predicted triples with respect to the training set as an important factor in interpreting results. We critically analyse the difficulty of mining novel commonsense knowledge, and show that a simple baseline method that outperforms the previous state of the art on predicting more novel triples.
2018-06-01
Proceedings of the Workshop on Generalization in the Age of Deep Learning (publié)
doi.org
arxiv.org