Call for Papers

List of Accepted Papers

Foveated Convolutions: Improving Spatial Transformer Networks by Modelling the Retina. Ethan Harris (University of Southampton)*; Mahesan Niranjan (University of Southampton); Jonathon Hare (University of Southampton)

Spatial and Colour Opponency in Anatomically Constrained Deep Networks. Ethan Harris (University of Southampton)*; Daniela Mihai (University of Southampton); Jonathon Hare (University of Southampton)

Shared Visual Abstractions. Tom White (Victoria University School of Design)*

Deep Prototype Models and Human Image Categorization. Pulkit Singh (Princeton University)*; Ruairidh Battleday (Princeton University); Joshua C Peterson (Princeton University); Thomas Griffiths (Princeton University)

Recurrent Architectures are Needed for Human-like Global Processing. Oh Hyeon Choung (EPFL)*; Adrien Doerig (EPFL); Alban Bornet (EPFL); Michael Herzog (EPFL)

Learning Non-Parametric Invariances from Data with Permanent Random Connectomes. Dipan K Pal (Carnegie Mellon University)*; Marios Savvides (Carnegie Mellon University)

V1Net: A computational model of long-range horizontal connections. Vijay Veerabadran (University of California San Diego)*; Virginia de Sa (UC San Diego)

To Decay or not to Decay: Modeling Video Memorability Over Time. Anelise Newman (MIT)*; Camilo L Fosco (MIT CSAIL); Aude Oliva (MIT); Vincent Casser (Harvard University); Barry McNamara (MIT)

How Many Glances? Modeling Multi-duration Saliency. Camilo L Fosco (MIT CSAIL)*; Anelise Newman (MIT); Pat Sukhum (Harvard); Yun Bin Zhang (Harvard); Aude Oliva (MIT); Zoya Bylinskii (Adobe Research)

Exploring CNN Inductive Biases: Shape vs. Texture. Katherine L. Hermann (Stanford University)*; Simon Kornblith (Google Brain)

Biologically-Motivated Deep Learning Method using Hierarchical Competitive Learning. Takashi Shinozaki (NICT CiNet)*

The Notorious Difficulty of Comparing Human and Machine Perception. Judy Borowski*, Christina M. Funke*, Karolina Stosio, Wieland Brendel, Thomas S. A. Wallis, Matthias Bethge. (University of Tuebingen)

When will AI misclassify? Human intuition for machine (mis)behavior. Zhenglong Zhou (University of Pennsylvania); Chaz Firestone (Johns Hopkins University)* .

The Emergence of Early Sound Categorical Responses in the Human Brain. Benjamin Lahner (MIT)*; Santani Teng (Smith-Kettlewell); Matthew Lowe (MIT); Ian Charest (University of Birmingham); Aude Oliva (MIT); Yalda Mohsenzadeh (The University of Western Ontario)

Bio-Inspired Hashing for Unsupervised Similarity Search. Chaitanya Ryali (UC San Diego)*; Dmitry Krotov (IBM Research AI)

Shared Representations of Stability in Humans, Supervised, & Unsupervised Neural Networks. Colin Conwell (Harvard University)*; Fenil Doshi (Harvard University); George Alvarez (Harvard University)

Approximating Human Judgment of Generated Image Quality. Y. Alex Kolchinski (Stanford University); Sharon Zhou (Stanford University)*; Shengjia Zhao (Stanford University); Mitchell L Gordon (Stanford University); Stefano Ermon (Stanford University)

Exploiting the modularity of deep networks to generate visual counterfactuals. Michel Besserve (MPI Tubingen)*; Arash Mehrjou (Mr.); Remy Sun (ENS Rennes); Bernhard Schölkopf (Max Planck Institute for Intelligent Systems)

Object Abstraction in Visual Model-Based Reinforcement Learning. John Co-Reyes (University of California Berkeley); Rishi Veerapaneni (UC Berkeley); Michael Chang (University of California, Berkeley)*; Michael Janner (UC Berkeley); Chelsea Finn (UC Berkeley); Jiajun Wu (MIT); Joshua Tenenbaum (MIT); Sergey Levine (UC Berkeley)

Update [Poster Reminders]

- There are no poster boards at workshops. Posters are taped to the wall with the special tabs that the NeurIPS staff needs to order.

- Please make your posters 36W x 48H inches or 90 x 122 cm.

- Posters should be on light weight paper, not laminated.


The following areas provide a sense of suitable topics for paper submissions:

  • Biological inspiration and inductive bias
  • Human-relevant strategies for robustness and generalization
  • New datasets (e.g., for comparing humans/animals and machines)
  • Biologically-driven self-supervision
  • Perceptual invariance and metamerism
  • Biologically-informed strategies to mitigate adversarial vulnerability
  • Foveation, active perception, and attention models
  • Intuitive physics
  • Perceptual and cognitive robustness
  • Nuances and noise in perceptual and cognitive Systems
  • Creative problem-solving
  • Differences and similarities between humans and deep neural networks
  • Canonical computations in biological and artificial systems
  • Alternative architectures for deep neural networks
  • Reverse engineering of the human visual system via deep neural networks

All submissions will be private and anonymous. Papers should be between 2-4 pages (excluding references), and will be formatted in NeurIPS style anonymously. Accepted papers will be presented as posters during the workshop, and will optionally be posted on the workshop website if the authors desire. Authors may optionally add appendixes in their submitted paper and the final submission including main paper, references and appendix should not exceed 10 pages. Supplementary Materials uploads are to only be used optionally for extra videos/code/data/figures and should be uploaded separately in the CMT submission website.

The paper submission process begins August 15st, 2019. Paper submission deadline is September 9th, 11th, 2019 11:59 pm PST. Link to Paper submission: https://cmt3.research.microsoft.com/SVRHM2019 . Authors can revise a submission multiple times before the final deadline, and are encouraged to update their submissions as many times as necessary before the deadline.

Papers accepted in this workshop will have non-archival status - thus researchers are encouraged to submit relevant on-going work in middle to late stages (or currently under review) in fields that span the domains of computer vision, machine learning, and vision science and cognitive science.

Papers that have been previously published in machine learning and computer vision venues (ex: CVPR, ICCV, ICLR, NeurIPS) should not be re-submitted as workshop papers -- though submitting ongoing work is highly encouraged. In addition, extending an abstract that has been previously accepted in other cognitive science and vision science venues that the machine learning community is not aware of such as: CCN, VSS, CogSci, SfN are highly encouraged! If such work is being re-submitted/extended please indicate so as a footnote in the abstract. Authors are allowed and encouraged to re-submit updated papers that did not get accepted at the main NeurIPS conference (or other computer vision/machine learning venue) that fits within the scope of this workshop.

Naturally, highly interdisciplinary papers that has never been presented before in any meeting/is not under review will receive higher consideration by the reviewers.

Here is the link for the Latex template style files for the submissions: https://neurips.cc/Conferences/2019/PaperInformation/StyleFiles

Please use this .sty template for the camera ready version: svrhm_2019.sty

All submitted papers will be evaluated using the following standards:

  1. Novelty of the idea with regards to human perception and cognition and how it may be relevant to modern machine learning.
  2. Rigor in preliminary theoretical contribution and/or empirical finding.
  3. Relevance at the intersection of the fields of machine learning and psychology.

NVIDIA will be donating a Titan RTX as best paper award

Oculus will be donating a Quest as best poster award

Reviewing Committee (Currently being updated):