Aidan Gomez

Toronto, Ontario, Canada Contact Info
23K followers 500+ connections

Join to view profile

About

I'm interested in making massive neural networks more efficient, and getting them…

Activity

Join now to see all activity

Experience & Education

  • Cohere

View Aidan’s full experience

By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.

Volunteer Experience

  • Good Shepherd Ministries Graphic

    Volunteer

    Good Shepherd Ministries

    - 9 months

    Poverty Alleviation

    The Good Shepherd is a homeless shelter in Toronto where hundreds line up every day to receive breakfast, lunch, and dinner. I had the honour of being able to serve these individuals breakfast and have conversations with my fellow Torontonians. The volunteers and staff that I worked alongside are stunning examples of human empathy, I'm endlessly grateful for what The Good Shepherd has given me.

  • Kawartha Pine Ridge District School Board Graphic

    Journey of Hope

    Kawartha Pine Ridge District School Board

    - 6 months

    Children

    The Journey of Hope is a humanitarian KPRDSB initiative sending students from three schools to Tanzania, Africa. Our roles there ranged from restoration of educational infrastructure, to educating students in computer skills. In addition we donated over 1200 pounds of supplies to various institutions across the country.

  • Volunteer

    The Companions of the Order of Malta, Oxford

    - Present 5 years 9 months

    Poverty Alleviation

    I've been incredibly fortunate to have been able to spend time having conversations with and serving food and drink to fellow Oxonians.

Publications

  • The Reversible Residual Network: Backpropagation Without Storing Activations

    Deep residual networks (ResNets) have significantly pushed forward the state-of-the-art on image classification, increasing in performance as networks grow both deeper and wider. However, memory consumption becomes a bottleneck, as one needs to store the activations in order to calculate gradients using backpropagation. We present the Reversible Residual Network (RevNet), a variant of ResNets where each layer's activations can be reconstructed exactly from the next layer's. Therefore, the…

    Deep residual networks (ResNets) have significantly pushed forward the state-of-the-art on image classification, increasing in performance as networks grow both deeper and wider. However, memory consumption becomes a bottleneck, as one needs to store the activations in order to calculate gradients using backpropagation. We present the Reversible Residual Network (RevNet), a variant of ResNets where each layer's activations can be reconstructed exactly from the next layer's. Therefore, the activations for most layers need not be stored in memory during backpropagation. We demonstrate the effectiveness of RevNets on CIFAR-10, CIFAR-100, and ImageNet, establishing nearly identical classification accuracy to equally-sized ResNets, even though the activation storage requirements are independent of depth.

    Other authors
    See publication

  • One Model To Learn Them All

    Arxiv

    Deep learning yields great results across many fields, from speech recognition, image classification, to translation. But for each problem, getting a deep model to work well involves research into the architecture and a long period of tuning. We present a single model that yields good results on a number of problems spanning multiple domains. In particular, this single model is trained concurrently on ImageNet, multiple translation tasks, image captioning (COCO dataset), a speech recognition…

    Deep learning yields great results across many fields, from speech recognition, image classification, to translation. But for each problem, getting a deep model to work well involves research into the architecture and a long period of tuning. We present a single model that yields good results on a number of problems spanning multiple domains. In particular, this single model is trained concurrently on ImageNet, multiple translation tasks, image captioning (COCO dataset), a speech recognition corpus, and an English parsing task. Our model architecture incorporates building blocks from multiple domains. It contains convolutional layers, an attention mechanism, and sparsely-gated layers. Each of these computational blocks is crucial for a subset of the tasks we train on. Interestingly, even if a block is not crucial for a task, we observe that adding it never hurts performance and in most cases improves it on all tasks. We also show that tasks with less data benefit largely from joint training with other tasks, while performance on large tasks degrades only slightly if at all.

    Other authors
    See publication

  • Attention Is All You Need

    The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more…

    The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.0 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.

    Other authors
    See publication

  • Depthwise Separable Convolutions for Neural Machine Translation

    Depthwise separable convolutions reduce the number of parameters and computation used in convolutional operations while increasing representational efficiency. They have been shown to be successful in image classification models, both in obtaining better models than previously possible for a given parameter count (the Xception architecture) and considerably reducing the number of parameters required to perform at a given level (the MobileNets family of architectures). Recently, convolutional…

    Depthwise separable convolutions reduce the number of parameters and computation used in convolutional operations while increasing representational efficiency. They have been shown to be successful in image classification models, both in obtaining better models than previously possible for a given parameter count (the Xception architecture) and considerably reducing the number of parameters required to perform at a given level (the MobileNets family of architectures). Recently, convolutional sequence-to-sequence networks have been applied to machine translation tasks with good results. In this work, we study how depthwise separable convolutions can be applied to neural machine translation. We introduce a new architecture inspired by Xception and ByteNet, called SliceNet, which enables a significant reduction of the parameter count and amount of computation needed to obtain results like ByteNet, and, with a similar parameter count, achieves new state-of-the-art results. In addition to showing that depthwise separable convolutions perform well for machine translation, we investigate the architectural changes that they enable: we observe that thanks to depthwise separability, we can increase the length of convolution windows, removing the need for filter dilation. We also introduce a new "super-separable" convolution operation that further reduces the number of parameters and computational cost for obtaining state-of-the-art results.

    Other authors
    See publication

  • Blog: The Neural Turing Machine

    A brief outline of the Neural Turing Machine's (NTM) design; a backpropogatable architecture that can (among many possibilities) learn to dynamically execute programs.

    See publication

  • Blog: Backpropogating an LSTM: A Numerical Example

    Medium

    LSTMs are arguably the most widely-used architecture in recurrent neural networks. This article walks through the mathematics behind these versatile units.

    See publication

  • Blog: Facebook on the creation of Machine Intelligence

    Medium

    An exploration of the technologies and philosophy being used to craft the first generation of artificial intelligence.

    See publication

Honors & Awards

  • AI Grant Fellow

    AI Grant

    aigrant.org - A fellowship sponsored by Google, CRV and others; started by Nat Friedman (Xamarin) and Daniel Gross (Y Combinator).

  • Clarendon Scholar

    -

    clarendon.ox.ac.uk - billed as Oxford’s most competitive graduate scholarship, the Clarendon scholarship is awarded exclusively based on academic performance and contribution.

  • Open Philanthropy AI Fellow

    Open Philanthropy

  • University College Alumni Scholar

    -

Languages

  • English

    Native or bilingual proficiency

More activity by Aidan

View Aidan’s full profile

  • See who you know in common
  • Get introduced
  • Contact Aidan directly
Join to view full profile

Other similar profiles

Explore more posts

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Others named Aidan Gomez

53 others named Aidan Gomez are on LinkedIn

See others named Aidan Gomez

Add new skills with these courses

See all courses