Crawl and Visualize ICLR 2020 OpenReview Data

Descriptions

This Jupyter Notebook contains the data crawled from ICLR 2020 OpenReview webpages and their visualizations. The list of submissions (sorted by the average ratings) can be found here.

Prerequisites

• Python3.6
• selenium
• pyvirtualdisplay (run on a headless device)
• numpy
• h5py
• matplotlib
• seaborn
• pandas
• imageio
• wordcloud

Visualizations

Decision

This Jupyter Notebook contains the data and visualizations that are crawled ICLR 2020 OpenReview webpages. All the crawled data (sorted by the average ratings) can be found here. The accepted papers have an average rating of 6.2431 and 3.4246 for rejected papers. The distribution is plotted as follows.

Rating distribution

The distribution of reviewer ratings centers around 4 (mean: 4.1837).

The cumulative sum of reviewer ratings.

You can compute how many papers are beaten by yours with

# See how many papers are beaten by yours
def PR(ratingmean, yourrating):
    pr = np.sum(yourrating > np.array(ratingmean))/len(rating_mean)*100
    samerating = np.sum(yourrating == np.array(ratingmean))/len(ratingmean)*100    
    return pr, same_rating
my_rating = (6+6+6)/3.  # your average rating here
pr, samerating = PR(ratingmean, my_rating)
print('Your papar ({:.2f}) is among the top {:.2f}% of submissions based on the ratings.\n'
      'There are {:.2f}% with the same rating.'.format(
          myrating, 100-pr, samerating))
           accept rate       orals    spotlight   posters
ICLR 2017: 39.1% (198/507)    15                    183
ICLR 2018: 32.0% (314/981)    23                    291
ICLR 2019: 31.4% (500/1591)   24                    476
ICLR 2020: 26.5% (687/2594)   48         108        529

[Output]

Your papar (6.00) is among the top 21.79% of submissions based on the ratings.
There are 8.24% with the same rating.

Word clouds

The word clouds formed by keywords of submissions show the hot topics including deep learning, reinforcement learning, representation learning, generative models, graph neural network, etc.

This figure is plotted with python word cloud generator

from wordcloud import WordCloud
wordcloud = WordCloud(maxfontsize=64, max_words=160, 
                      width=1280, height=640,
                      background_color="black").generate(' '.join(keywords))
plt.figure(figsize=(16, 8))
plt.imshow(wordcloud, interpolation="bilinear")
plt.axis("off")
plt.show()

Frequent keywords

The top 50 common keywords and their frequency.

The average reviewer ratings and the frequency of keywords indicate that to maximize your chance to get higher ratings would be using the keywords such as compositionality, deep learning theory, or gradient descent.

Review length histogram

The average review length is 407.91 words. The histogram is as follows.

Reviewer rating change during the rebuttal period

All individual ratings:

The average rating for each paper:

Top authors

The authors with more than 5 submissions.

How it works

See How to install Selenium and ChromeDriver on Ubuntu.

To crawl data from dynamic websites such as OpenReview, a headless web simulator can be created by

from selenium import webdriver
from selenium.webdriver.chrome.options import Options
executable_path = '/Users/waltersun/Desktop/chromedriver'  # path to your executable browser
options = Options()
options.add_argument("--headless")
browser = webdriver.Chrome(options=options, executablepath=executablepath)

Then, we can get the content from a webpage

browser.get(url)

To know what content we to crawl, we need to inspect the webpage layout.

I chose to get the content by

key = browser.findelementsbyclassname("notecontentfield")
value = browser.findelementsbyclassname("notecontentvalue")

The data includes the abstract, keywords, TL; DR, comments.

Installing Selenium and ChromeDriver on Ubuntu

The following content is hugely borrowed from a nice post written by Christopher Su.

• Install Google Chrome for Debian/Ubuntu

sudo apt-get install libxss1 libappindicator1 libindicator7
wget https://dl.google.com/linux/direct/google-chrome-stablecurrentamd64.deb
sudo dpkg -i google-chrome*.deb sudo apt-get install -f

• Install xvfb to run Chrome on a headless device

sudo apt-get install xvfb

• Install ChromeDriver for 64-bit Linux

sudo apt-get install unzip  # If you don't have unzip package
wget -N http://chromedriver.storage.googleapis.com/2.26/chromedriver_linux64.zip unzip chromedriver_linux64.zip chmod +x chromedriver
sudo mv -f chromedriver /usr/local/share/chromedriver sudo ln -s /usr/local/share/chromedriver /usr/local/bin/chromedriver sudo ln -s /usr/local/share/chromedriver /usr/bin/chromedriver

If your system is 32-bit, please find the ChromeDriver releases here and modify the above download command.

• Install Python dependencies (Selenium and pyvirtualdisplay)

pip install pyvirtualdisplay selenium

• Test your setup in Python

from pyvirtualdisplay import Display
from selenium import webdriver
display = Display(visible=0, size=(1024, 1024)) display.start() browser = webdriver.Chrome() browser.get('http://shaohua0116.github.io/') print(browser.title) print(browser.findelementbyclassname('bio').text)

All ICLR 2020 OpenReview data

Collected at 12/23/2019 03:59:42 PM

Number of submissions: 2594 (withdrawn/desk reject submissions: 383)

| Rank | Average Rating | Title | Ratings | Variance | Decision | | :---: | :---: | :--- | :---: | :---: | :---: | | 1 | 8.00 | Hyper-sagnn: A Self-attention Based Graph Neural Network For Hypergraphs | 8, 8 | 0.00 | Accept (Poster) | | 2 | 8.00 | Freelb: Enhanced Adversarial Training For Language Understanding | 8, 8 | 0.00 | Accept (Spotlight) | | 3 | 8.00 | Enhancing Adversarial Defense By K-winners-take-all | 8, 8, 8 | 0.00 | Accept (Spotlight) | | 4 | 8.00 | Implementation Matters In Deep Rl: A Case Study On Ppo And Trpo | 8, 8, 8 | 0.00 | Accept (Talk) | | 5 | 8.00 | Contrastive Learning Of Structured World Models | 8, 8, 8 | 0.00 | Accept (Talk) | | 6 | 8.00 | Learning To Balance: Bayesian Meta-learning For Imbalanced And Out-of-distribution Tasks | 8, 8, 8 | 0.00 | Accept (Talk) | | 7 | 8.00 | Sparse Coding With Gated Learned Ista | 8, 8, 8 | 0.00 | Accept (Spotlight) | | 8 | 8.00 | Restricting The Flow: Information Bottlenecks For Attribution | 8, 8, 8 | 0.00 | Accept (Talk) | | 9 | 8.00 | Causal Discovery With Reinforcement Learning | 8, 8, 8 | 0.00 | Accept (Talk) | | 10 | 8.00 | Dynamics-aware Unsupervised Skill Discovery | 8, 8, 8 | 0.00 | Accept (Talk) | | 11 | 8.00 | Nas-bench-102: Extending The Scope Of Reproducible Neural Architecture Search | 8, 8, 8 | 0.00 | Accept (Spotlight) | | 12 | 8.00 | Data-dependent Gaussian Prior Objective For Language Generation | 8, 8, 8 | 0.00 | Accept (Talk) | | 13 | 8.00 | Gendice: Generalized Offline Estimation Of Stationary Values | 8, 8, 8 | 0.00 | Accept (Talk) | | 14 | 8.00 | Mathematical Reasoning In Latent Space | 8, 8, 8 | 0.00 | Accept (Talk) | | 15 | 8.00 | Why Gradient Clipping Accelerates Training: A Theoretical Justification For Adaptivity | 8, 8, 8 | 0.00 | Accept (Talk) | | 16 | 8.00 | Cater: A Diagnostic Dataset For Compositional Actions & Temporal Reasoning | 8, 8, 8 | 0.00 | Accept (Talk) | | 17 | 8.00 | Understanding And Robustifying Differentiable Architecture Search | 8, 8, 8 | 0.00 | Accept (Talk) | | 18 | 8.00 | Geometric Analysis Of Nonconvex Optimization Landscapes For Overcomplete Learning | 8, 8, 8 | 0.00 | Accept (Talk) | | 19 | 8.00 | Simplified Action Decoder For Deep Multi-agent Reinforcement Learning | 8, 8, 8 | 0.00 | Accept (Spotlight) | | 20 | 8.00 | Mirror-generative Neural Machine Translation | 8, 8, 8 | 0.00 | Accept (Talk) | | 21 | 8.00 | On The "steerability" Of Generative Adversarial Networks | 8, 8, 8 | 0.00 | Accept (Poster) | | 22 | 8.00 | A Theory Of Usable Information Under Computational Constraints | 8, 8 | 0.00 | Accept (Talk) | | 23 | 8.00 | How Much Position Information Do Convolutional Neural Networks Encode? | 8, 8, 8 | 0.00 | Accept (Spotlight) | | 24 | 8.00 | Principled Weight Initialization For Hypernetworks | 8, 8, 8 | 0.00 | Accept (Talk) | | 25 | 8.00 | Meta-learning With Warped Gradient Descent | 8, 8, 8 | 0.00 | Accept (Talk) | | 26 | 8.00 | Rotation-invariant Clustering Of Functional Cell Types In Primary Visual Cortex | 8, 8, 8 | 0.00 | Accept (Talk) | | 27 | 8.00 | Depth-width Trade-offs For Relu Networks Via Sharkovsky's Theorem | 8, 8 | 0.00 | Accept (Spotlight) | | 28 | 8.00 | The Logical Expressiveness Of Graph Neural Networks | 8, 8, 8 | 0.00 | Accept (Spotlight) | | 29 | 8.00 | Differentiation Of Blackbox Combinatorial Solvers | 8, 8, 8 | 0.00 | Accept (Spotlight) | | 30 | 8.00 | A Generalized Training Approach For Multiagent Learning | 8, 8, 8 | 0.00 | Accept (Talk) | | 31 | 8.00 | Smooth Markets: A Basic Mechanism For Organizing Gradient-based Learners | 8, 8 | 0.00 | Accept (Poster) | | 32 | 8.00 | Backpack: Packing More Into Backprop | 8, 8, 8 | 0.00 | Accept (Talk) | | 33 | 8.00 | Differentiable Reasoning Over A Virtual Knowledge Base | 8, 8, 8 | 0.00 | Accept (Talk) | | 34 | 8.00 | Optimal Strategies Against Generative Attacks | 8, 8, 8, 8 | 0.00 | Accept (Talk) | | 35 | 7.50 | Vq-wav2vec: Self-supervised Learning Of Discrete Speech Representations | 8, 6, 8, 8 | 0.87 | Accept (Poster) | | 36 | 7.50 | Rna Secondary Structure Prediction By Learning Unrolled Algorithms | 8, 8, 8, 6 | 0.87 | Accept (Talk) | | 37 | 7.33 | Doubly Robust Bias Reduction In Infinite Horizon Off-policy Estimation | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 38 | 7.33 | Meta-learning Without Memorization | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 39 | 7.33 | Directional Message Passing For Molecular Graphs | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 40 | 7.33 | Learning Robust Representations Via Multi-view Information Bottleneck | 6, 8, 8 | 0.94 | Accept (Poster) | | 41 | 7.33 | Polylogarithmic Width Suffices For Gradient Descent To Achieve Arbitrarily Small Test Error With Shallow Relu Networks | 8, 6, 8 | 0.94 | Accept (Poster) | | 42 | 7.33 | Mixed-curvature Variational Autoencoders | 6, 8, 8 | 0.94 | Accept (Poster) | | 43 | 7.33 | Federated Learning With Matched Averaging | 6, 8, 8 | 0.94 | Accept (Talk) | | 44 | 7.33 | Deep Network Classification By Scattering And Homotopy Dictionary Learning | 8, 8, 6 | 0.94 | Accept (Poster) | | 45 | 7.33 | Finite Depth And Width Corrections To The Neural Tangent Kernel | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 46 | 7.33 | A Closer Look At Deep Policy Gradients | 8, 6, 8 | 0.94 | Accept (Talk) | | 47 | 7.33 | Measuring The Reliability Of Reinforcement Learning Algorithms | 8, 8, 6 | 0.94 | Accept (Spotlight) | | 48 | 7.33 | Compressive Transformers For Long-range Sequence Modelling | 6, 8, 8 | 0.94 | Accept (Poster) | | 49 | 7.33 | Truth Or Backpropaganda? An Empirical Investigation Of Deep Learning Theory | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 50 | 7.33 | Fasterseg: Searching For Faster Real-time Semantic Segmentation | 6, 8, 8 | 0.94 | Accept (Poster) | | 51 | 7.33 | Classification-based Anomaly Detection For General Data | 8, 8, 6 | 0.94 | Accept (Poster) | | 52 | 7.33 | Robust Subspace Recovery Layer For Unsupervised Anomaly Detection | 6, 8, 8 | 0.94 | Accept (Poster) | | 53 | 7.33 | At Stability's Edge: How To Adjust Hyperparameters To Preserve Minima Selection In Asynchronous Training Of Neural Networks? | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 54 | 7.33 | Albert: A Lite Bert For Self-supervised Learning Of Language Representations | 8, 8, 6 | 0.94 | Accept (Spotlight) | | 55 | 7.33 | On Mutual Information Maximization For Representation Learning | 8, 8, 6 | 0.94 | Accept (Poster) | | 56 | 7.33 | Deep Imitative Models For Flexible Inference, Planning, And Control | 8, 6, 8 | 0.94 | Accept (Poster) | | 57 | 7.33 | Reconstructing Continuous Distributions Of 3d Protein Structure From Cryo-em Images | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 58 | 7.33 | On The Convergence Of Fedavg On Non-iid Data | 6, 8, 8 | 0.94 | Accept (Talk) | | 59 | 7.33 | Comparing Fine-tuning And Rewinding In Neural Network Pruning | 8, 6, 8 | 0.94 | Accept (Talk) | | 60 | 7.33 | Low-resource Knowledge-grounded Dialogue Generation | 6, 8, 8 | 0.94 | Accept (Poster) | | 61 | 7.33 | Network Deconvolution | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 62 | 7.33 | A Mutual Information Maximization Perspective Of Language Representation Learning | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 63 | 7.33 | Lambdanet: Probabilistic Type Inference Using Graph Neural Networks | 6, 8, 8 | 0.94 | Accept (Poster) | | 64 | 7.33 | On The Equivalence Between Node Embeddings And Structural Graph Representations | 6, 8, 8 | 0.94 | Accept (Poster) | | 65 | 7.33 | Intensity-free Learning Of Temporal Point Processes | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 66 | 7.33 | Neural Network Branching For Neural Network Verification | 8, 6, 8 | 0.94 | Accept (Talk) | | 67 | 7.33 | Graphzoom: A Multi-level Spectral Approach For Accurate And Scalable Graph Embedding | 8, 8, 6 | 0.94 | Accept (Talk) | | 68 | 7.33 | Adversarial Training And Provable Defenses: Bridging The Gap | 8, 6, 8 | 0.94 | Accept (Talk) | | 69 | 7.33 | Harnessing The Power Of Infinitely Wide Deep Nets On Small-data Tasks | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 70 | 7.33 | Online And Stochastic Optimization Beyond Lipschitz Continuity: A Riemannian Approach | 8, 8, 6 | 0.94 | Accept (Spotlight) | | 71 | 7.33 | Meta-q-learning | 8, 8, 6 | 0.94 | Accept (Talk) | | 72 | 7.33 | Symplectic Ode-net: Learning Hamiltonian Dynamics With Control | 6, 8, 8 | 0.94 | Accept (Poster) | | 73 | 7.33 | Electra: Pre-training Text Encoders As Discriminators Rather Than Generators | 8, 8, 6 | 0.94 | Accept (Poster) | | 74 | 7.33 | The Ingredients Of Real World Robotic Reinforcement Learning | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 75 | 7.33 | Generalization Of Two-layer Neural Networks: An Asymptotic Viewpoint | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 76 | 7.33 | Watch The Unobserved: A Simple Approach To Parallelizing Monte Carlo Tree Search | 8, 6, 8 | 0.94 | Accept (Talk) | | 77 | 7.33 | Fspool: Learning Set Representations With Featurewise Sort Pooling | 8, 8, 6 | 0.94 | Accept (Poster) | | 78 | 7.33 | Seed Rl: Scalable And Efficient Deep-rl With Accelerated Central Inference | 8, 6, 8 | 0.94 | Accept (Talk) | | 79 | 7.33 | Fast Task Inference With Variational Intrinsic Successor Features | 8, 6, 8 | 0.94 | Accept (Talk) | | 80 | 7.33 | Stable Rank Normalization For Improved Generalization In Neural Networks And Gans | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 81 | 7.33 | Ddsp: Differentiable Digital Signal Processing | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 82 | 7.33 | Deep Batch Active Learning By Diverse, Uncertain Gradient Lower Bounds | 8, 6, 8 | 0.94 | Accept (Talk) | | 83 | 7.33 | Massively Multilingual Sparse Word Representations | 6, 8, 8 | 0.94 | Accept (Poster) | | 84 | 7.33 | Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps | 8, 8, 6 | 0.94 | Accept (Spotlight) | | 85 | 7.33 | Mogrifier Lstm | 6, 8, 8 | 0.94 | Accept (Talk) | | 86 | 7.33 | Scaling Autoregressive Video Models | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 87 | 7.33 | Meta-learning Acquisition Functions For Transfer Learning In Bayesian Optimization | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 88 | 7.33 | Latent Normalizing Flows For Many-to-many Cross Domain Mappings | 6, 8, 8 | 0.94 | Accept (Poster) | | 89 | 7.33 | Cyclical Stochastic Gradient Mcmc For Bayesian Deep Learning | 6, 8, 8 | 0.94 | Accept (Talk) | | 90 | 7.33 | Discriminative Particle Filter Reinforcement Learning For Complex Partial Observations | 8, 6, 8 | 0.94 | Accept (Poster) | | 91 | 7.33 | Deep Learning For Symbolic Mathematics | 8, 8, 6 | 0.94 | Accept (Spotlight) | | 92 | 7.33 | What Graph Neural Networks Cannot Learn: Depth Vs Width | 8, 6, 8 | 0.94 | Accept (Poster) | | 93 | 7.33 | Sumo: Unbiased Estimation Of Log Marginal Probability For Latent Variable Models | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 94 | 7.33 | Ted: A Pretrained Unsupervised Summarization Model With Theme Modeling And Denoising | 6, 8, 8 | 0.94 | Reject | | 95 | 7.33 | Program Guided Agent | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 96 | 7.33 | Your Classifier Is Secretly An Energy Based Model And You Should Treat It Like One | 6, 8, 8 | 0.94 | Accept (Talk) | | 97 | 7.33 | Disentangling Neural Mechanisms For Perceptual Grouping | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 98 | 7.33 | Symplectic Recurrent Neural Networks | 8, 8, 6 | 0.94 | Accept (Spotlight) | | 99 | 7.33 | Assemblenet: Searching For Multi-stream Neural Connectivity In Video Architectures | 6, 8, 8 | 0.94 | Accept (Poster) | | 100 | 7.33 | Harnessing Structures For Value-based Planning And Reinforcement Learning | 6, 8, 8 | 0.94 | Accept (Talk) | | 101 | 7.33 | When Do Variational Autoencoders Know What They Don't Know? | 6, 8, 8 | 0.94 | N/A | | 102 | 7.33 | Physics-aware Difference Graph Networks For Sparsely-observed Dynamics | 8, 8, 6 | 0.94 | Accept (Poster) | | 103 | 7.33 | Observational Overfitting In Reinforcement Learning | 6, 8, 8 | 0.94 | Accept (Poster) | | 104 | 7.33 | Learning To Plan In High Dimensions Via Neural Exploration-exploitation Trees | 8, 8, 6 | 0.94 | Accept (Spotlight) | | 105 | 7.33 | Cross-lingual Alignment Vs Joint Training: A Comparative Study And A Simple Unified Framework | 6, 8, 8 | 0.94 | Accept (Poster) | | 106 | 7.33 | Unbiased Contrastive Divergence Algorithm For Training Energy-based Latent Variable Models | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 107 | 7.33 | Explain Your Move: Understanding Agent Actions Using Focused Feature Saliency | 6, 8, 8 | 0.94 | Accept (Poster) | | 108 | 7.33 | Glad: Learning Sparse Graph Recovery | 8, 6, 8 | 0.94 | Accept (Poster) | | 109 | 7.33 | What Can Neural Networks Reason About? | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 110 | 7.33 | End To End Trainable Active Contours Via Differentiable Rendering | 8, 8, 6 | 0.94 | Accept (Poster) | | 111 | 7.33 | High Fidelity Speech Synthesis With Adversarial Networks | 8, 6, 8 | 0.94 | Accept (Talk) | | 112 | 7.33 | Thieves On Sesame Street! Model Extraction Of Bert-based Apis | 6, 8, 8 | 0.94 | Accept (Poster) | | 113 | 7.33 | Convolutional Conditional Neural Processes | 6, 8, 8 | 0.94 | Accept (Talk) | | 114 | 7.33 | Is A Good Representation Sufficient For Sample Efficient Reinforcement Learning? | 8, 8, 6 | 0.94 | Accept (Spotlight) | | 115 | 7.33 | Graph Neural Networks Exponentially Lose Expressive Power For Node Classification | 8, 6, 8 | 0.94 | Accept (Spotlight) | | 116 | 7.33 | Learning Hierarchical Discrete Linguistic Units From Visually-grounded Speech | 6, 8, 8 | 0.94 | Accept (Talk) | | 117 | 7.33 | Poly-encoders: Architectures And Pre-training Strategies For Fast And Accurate Multi-sentence Scoring | 8, 6, 8 | 0.94 | Accept (Poster) | | 118 | 7.33 | Progressive Learning And Disentanglement Of Hierarchical Representations | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 119 | 7.33 | Gradient Descent Maximizes The Margin Of Homogeneous Neural Networks | 8, 8, 6 | 0.94 | Accept (Talk) | | 120 | 7.33 | Energy-based Models For Atomic-resolution Protein Conformations | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 121 | 7.33 | Disagreement-regularized Imitation Learning | 6, 8, 8 | 0.94 | Accept (Spotlight) | | 122 | 7.33 | Sequential Latent Knowledge Selection For Knowledge-grounded Dialogue | 8, 8, 6 | 0.94 | Accept (Spotlight) | | 123 | 7.33 | Reformer: The Efficient Transformer | 8, 8, 6 | 0.94 | Accept (Talk) | | 124 | 7.00 | Target-embedding Autoencoders For Supervised Representation Learning | 6, 8, 6, 8 | 1.00 | Accept (Talk) | | 125 | 7.00 | Memo: A Deep Network For Flexible Combination Of Episodic Memories | 6, 8 | 1.00 | Accept (Poster) | | 126 | 7.00 | Neural Tangent Kernels, Transportation Mappings, And Universal Approximation | 8, 6 | 1.00 | Accept (Poster) | | 127 | 7.00 | Sliced Cramer Synaptic Consolidation For Preserving Deeply Learned Representations | 6, 8 | 1.00 | Accept (Spotlight) | | 128 | 7.00 | Encoding Word Order In Complex Embeddings | 8, 6, 8, 6 | 1.00 | Accept (Spotlight) | | 129 | 7.00 | An Exponential Learning Rate Schedule For Batch Normalized Networks | 8, 8, 6, 6 | 1.00 | Accept (Spotlight) | | 130 | 7.00 | Spectral Embedding Of Regularized Block Models | 8, 6 | 1.00 | Accept (Spotlight) | | 131 | 7.00 | How The Choice Of Activation Affects Training Of Overparametrized Neural Nets | 6, 8 | 1.00 | Accept (Poster) | | 132 | 7.00 | Double Neural Counterfactual Regret Minimization | 8, 6 | 1.00 | Accept (Poster) | | 133 | 7.00 | Building Deep Equivariant Capsule Networks | 8, 6 | 1.00 | Accept (Talk) | | 134 | 7.00 | Ridge Regression: Structure, Cross-validation, And Sketching | 6, 8 | 1.00 | Accept (Spotlight) | | 135 | 7.00 | Quantum Algorithms For Deep Convolutional Neural Networks | 6, 8, 8, 6 | 1.00 | Accept (Poster) | | 136 | 7.00 | Biologically Inspired Sleep Algorithm For Increased Generalization And Adversarial Robustness In Deep Neural Networks | 6, 8 | 1.00 | Accept (Poster) | | 137 | 7.00 | And The Bit Goes Down: Revisiting The Quantization Of Neural Networks | 8, 6, 8, 6 | 1.00 | Accept (Spotlight) | | 138 | 7.00 | Dream To Control: Learning Behaviors By Latent Imagination | 8, 6, 6, 8 | 1.00 | Accept (Spotlight) | | 139 | 7.00 | Understanding L4-based Dictionary Learning: Interpretation, Stability, And Robustness | 8, 6 | 1.00 | Accept (Poster) | | 140 | 7.00 | Language Gans Falling Short | 6, 8 | 1.00 | Accept (Poster) | | 141 | 7.00 | Explanation By Progressive Exaggeration | 6, 8 | 1.00 | Accept (Spotlight) | | 142 | 6.75 | An Inductive Bias For Distances: Neural Nets That Respect The Triangle Inequality | 8, 8, 3, 8 | 2.17 | Accept (Poster) | | 143 | 6.67 | Fsnet: Compression Of Deep Convolutional Neural Networks By Filter Summary | 8, 6, 6 | 0.94 | Accept (Poster) | | 144 | 6.67 | Neural Outlier Rejection For Self-supervised Keypoint Learning | 6, 6, 8 | 0.94 | Accept (Poster) | | 145 | 6.67 | On Robustness Of Neural Ordinary Differential Equations | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 146 | 6.67 | Controlling Generative Models With Continuous Factors Of Variations | 6, 8, 6 | 0.94 | Accept (Poster) | | 147 | 6.67 | A Latent Morphology Model For Open-vocabulary Neural Machine Translation | 8, 6, 6 | 0.94 | Accept (Spotlight) | | 148 | 6.67 | Are Pre-trained Language Models Aware Of Phrases? Simple But Strong Baselines For Grammar Induction | 6, 6, 8 | 0.94 | Accept (Poster) | | 149 | 6.67 | Continual Learning With Hypernetworks | 6, 8, 6 | 0.94 | Accept (Poster) | | 150 | 6.67 | The Function Of Contextual Illusions | 6, 6, 8 | 0.94 | Accept (Poster) | | 151 | 6.67 | Training Individually Fair Ml Models With Sensitive Subspace Robustness | 8, 6, 6 | 0.94 | Accept (Spotlight) | | 152 | 6.67 | Estimating Gradients For Discrete Random Variables By Sampling Without Replacement | 6, 6, 8 | 0.94 | Accept (Spotlight) | | 153 | 6.67 | Reinforced Genetic Algorithm Learning For Optimizing Computation Graphs | 8, 6, 6 | 0.94 | Accept (Poster) | | 154 | 6.67 | Asymptotics Of Wide Networks From Feynman Diagrams | 8, 6, 6 | 0.94 | Accept (Spotlight) | | 155 | 6.67 | Gradient-based Neural Dag Learning | 6, 6, 8 | 0.94 | Accept (Poster) | | 156 | 6.67 | Query-efficient Meta Attack To Deep Neural Networks | 6, 8, 6 | 0.94 | Accept (Poster) | | 157 | 6.67 | Padé Activation Units: End-to-end Learning Of Flexible Activation Functions In Deep Networks | 6, 8, 6 | 0.94 | Accept (Poster) | | 158 | 6.67 | The Intriguing Role Of Module Criticality In The Generalization Of Deep Networks | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 159 | 6.67 | Intrinsically Motivated Discovery Of Diverse Patterns In Self-organizing Systems | 6, 6, 8 | 0.94 | Accept (Talk) | | 160 | 6.67 | Rényi Fair Inference | 6, 6, 8 | 0.94 | Accept (Poster) | | 161 | 6.67 | Breaking Certified Defenses: Semantic Adversarial Examples With Spoofed Robustness Certificates | 6, 8, 6 | 0.94 | Accept (Poster) | | 162 | 6.67 | Influence-based Multi-agent Exploration | 6, 6, 8 | 0.94 | Accept (Spotlight) | | 163 | 6.67 | Emergence Of Functional And Structural Properties Of The Head Direction System By Optimization Of Recurrent Neural Networks | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 164 | 6.67 | Lipschitz Constant Estimation For Neural Networks Via Sparse Polynomial Optimization | 6, 8, 6 | 0.94 | Accept (Poster) | | 165 | 6.67 | Monotonic Multihead Attention | 6, 8, 6 | 0.94 | Accept (Poster) | | 166 | 6.67 | Amrl: Aggregated Memory For Reinforcement Learning | 6, 6, 8 | 0.94 | Accept (Poster) | | 167 | 6.67 | Deepsphere: A Graph-based Spherical Cnn | 8, 6, 6 | 0.94 | Accept (Spotlight) | | 168 | 6.67 | On Identifiability In Transformers | 6, 8, 6 | 0.94 | Accept (Poster) | | 169 | 6.67 | Semi-supervised Generative Modeling For Controllable Speech Synthesis | 6, 8, 6 | 0.94 | Accept (Poster) | | 170 | 6.67 | Reclor: A Reading Comprehension Dataset Requiring Logical Reasoning | 6, 6, 8 | 0.94 | Accept (Poster) | | 171 | 6.67 | Learning To Control Pdes With Differentiable Physics | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 172 | 6.67 | Hamiltonian Generative Networks | 8, 6, 6 | 0.94 | Accept (Spotlight) | | 173 | 6.67 | Intrinsic Motivation For Encouraging Synergistic Behavior | 6, 8, 6 | 0.94 | Accept (Poster) | | 174 | 6.67 | Fast Is Better Than Free: Revisiting Adversarial Training | 8, 6, 6 | 0.94 | Accept (Poster) | | 175 | 6.67 | Where Is The Information In A Deep Network? | 6, 8, 6 | 0.94 | Reject | | 176 | 6.67 | A Fair Comparison Of Graph Neural Networks For Graph Classification | 6, 8, 6 | 0.94 | Accept (Poster) | | 177 | 6.67 | Co-attentive Equivariant Neural Networks: Focusing Equivariance On Transformations Co-occurring In Data | 6, 8, 6 | 0.94 | Accept (Poster) | | 178 | 6.67 | Robust Reinforcement Learning For Continuous Control With Model Misspecification | 6, 6, 8 | 0.94 | Accept (Poster) | | 179 | 6.67 | Safe Policy Learning For Continuous Control | 6, 8, 6 | 0.94 | Reject | | 180 | 6.67 | Permutation Equivariant Models For Compositional Generalization In Language | 8, 6, 6 | 0.94 | Accept (Poster) | | 181 | 6.67 | Estimating Counterfactual Treatment Outcomes Over Time Through Adversarially Balanced Representations | 8, 6, 6 | 0.94 | Accept (Spotlight) | | 182 | 6.67 | Single Path One-shot Neural Architecture Search With Uniform Sampling | 6, 6, 8 | 0.94 | Reject | | 183 | 6.67 | Learning To Retrieve Reasoning Paths Over Wikipedia Graph For Question Answering | 6, 6, 8 | 0.94 | Accept (Poster) | | 184 | 6.67 | Learning To Anneal And Prune Proximity Graphs For Similarity Search | 6, 6, 8 | 0.94 | Reject | | 185 | 6.67 | Evolutionary Population Curriculum For Scaling Multi-agent Reinforcement Learning | 6, 8, 6 | 0.94 | Accept (Poster) | | 186 | 6.67 | Sqil: Imitation Learning Via Reinforcement Learning With Sparse Rewards | 8, 6, 6 | 0.94 | Accept (Poster) | | 187 | 6.67 | Never Give Up: Learning Directed Exploration Strategies | 6, 6, 8 | 0.94 | Accept (Poster) | | 188 | 6.67 | On The Interaction Between Supervision And Self-play In Emergent Communication | 6, 8, 6 | 0.94 | Accept (Poster) | | 189 | 6.67 | Simple And Effective Regularization Methods For Training On Noisily Labeled Data With Generalization Guarantee | 6, 8, 6 | 0.94 | Accept (Poster) | | 190 | 6.67 | Learning To Learn Kernels With Variational Random Features | 8, 6, 6 | 0.94 | Reject | | 191 | 6.67 | Locality And Compositionality In Zero-shot Learning | 8, 6, 6 | 0.94 | Accept (Poster) | | 192 | 6.67 | Extreme Tensoring For Low-memory Preconditioning | 8, 6, 6 | 0.94 | Accept (Poster) | | 193 | 6.67 | Towards Stabilizing Batch Statistics In Backward Propagation Of Batch Normalization | 6, 8, 6 | 0.94 | Accept (Poster) | | 194 | 6.67 | Distributed Bandit Learning: Near-optimal Regret With Efficient Communication | 8, 6, 6 | 0.94 | Accept (Poster) | | 195 | 6.67 | Clevrer: Collision Events For Video Representation And Reasoning | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 196 | 6.67 | Diverse Trajectory Forecasting With Determinantal Point Processes | 8, 6, 6 | 0.94 | Accept (Poster) | | 197 | 6.67 | Decoupling Representation And Classifier For Long-tailed Recognition | 6, 8, 6 | 0.94 | Accept (Poster) | | 198 | 6.67 | Mutual Exclusivity As A Challenge For Deep Neural Networks | 6, 8, 6 | 0.94 | Reject | | 199 | 6.67 | Scalable Model Compression By Entropy Penalized Reparameterization | 6, 8, 6 | 0.94 | Accept (Poster) | | 200 | 6.67 | Snode: Spectral Discretization Of Neural Odes For System Identification | 6, 6, 8 | 0.94 | Accept (Poster) | | 201 | 6.67 | Learning Expensive Coordination: An Event-based Deep Rl Approach | 6, 8, 6 | 0.94 | Accept (Poster) | | 202 | 6.67 | You Can Teach An Old Dog New Tricks! On Training Knowledge Graph Embeddings | 8, 6, 6 | 0.94 | Accept (Poster) | | 203 | 6.67 | Synthesizing Programmatic Policies That Inductively Generalize | 6, 8, 6 | 0.94 | Accept (Poster) | | 204 | 6.67 | Denoising And Regularization Via Exploiting The Structural Bias Of Convolutional Generators | 6, 8, 6 | 0.94 | Accept (Poster) | | 205 | 6.67 | Incremental Rnn: A Dynamical View. | 8, 6, 6 | 0.94 | Accept (Poster) | | 206 | 6.67 | Tabfact: A Large-scale Dataset For Table-based Fact Verification | 8, 6, 6 | 0.94 | Accept (Poster) | | 207 | 6.67 | Multiplicative Interactions And Where To Find Them | 6, 8, 6 | 0.94 | Accept (Poster) | | 208 | 6.67 | U-gat-it: Unsupervised Generative Attentional Networks With Adaptive Layer-instance Normalization For Image-to-image Translation | 6, 8, 6 | 0.94 | Accept (Poster) | | 209 | 6.67 | Making Sense Of Reinforcement Learning And Probabilistic Inference | 6, 6, 8 | 0.94 | Accept (Spotlight) | | 210 | 6.67 | Improving Adversarial Robustness Requires Revisiting Misclassified Examples | 8, 6, 6 | 0.94 | Accept (Poster) | | 211 | 6.67 | Learning To Learn By Zeroth-order Oracle | 6, 8, 6 | 0.94 | Accept (Poster) | | 212 | 6.67 | Query2box: Reasoning Over Knowledge Graphs In Vector Space Using Box Embeddings | 6, 6, 8 | 0.94 | Accept (Poster) | | 213 | 6.67 | Deep Double Descent: Where Bigger Models And More Data Hurt | 8, 6, 6 | 0.94 | Accept (Poster) | | 214 | 6.67 | Training Generative Adversarial Networks From Incomplete Observations Using Factorised Discriminators | 6, 8, 6 | 0.94 | Accept (Poster) | | 215 | 6.67 | Consistency Regularization For Generative Adversarial Networks | 8, 6, 6 | 0.94 | Accept (Poster) | | 216 | 6.67 | Sign Bits Are All You Need For Black-box Attacks | 8, 6, 6 | 0.94 | Accept (Poster) | | 217 | 6.67 | Inductive Representation Learning On Temporal Graphs | 6, 6, 8 | 0.94 | Accept (Poster) | | 218 | 6.67 | Neural Symbolic Reader: Scalable Integration Of Distributed And Symbolic Representations For Reading Comprehension | 6, 6, 8 | 0.94 | Accept (Spotlight) | | 219 | 6.67 | Decoding As Dynamic Programming For Recurrent Autoregressive Models | 6, 6, 8 | 0.94 | Accept (Poster) | | 220 | 6.67 | Neural Module Networks For Reasoning Over Text | 6, 8, 6 | 0.94 | Accept (Poster) | | 221 | 6.67 | Multi-agent Interactions Modeling With Correlated Policies | 6, 6, 8 | 0.94 | Accept (Poster) | | 222 | 6.67 | Actor-critic Provably Finds Nash Equilibria Of Linear-quadratic Mean-field Games | 6, 6, 8 | 0.94 | Accept (Poster) | | 223 | 6.67 | Scale-equivariant Steerable Networks | 6, 6, 8 | 0.94 | Accept (Poster) | | 224 | 6.67 | Kernelized Wasserstein Natural Gradient | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 225 | 6.67 | Batch-shaping For Learning Conditional Channel Gated Networks | 6, 6, 8 | 0.94 | Accept (Poster) | | 226 | 6.67 | Intriguing Properties Of Adversarial Training At Scale | 6, 8, 6 | 0.94 | Accept (Poster) | | 227 | 6.67 | Improving Generalization In Meta Reinforcement Learning Using Neural Objectives | 6, 6, 8 | 0.94 | Accept (Spotlight) | | 228 | 6.67 | Spike-based Causal Inference For Weight Alignment | 8, 6, 6 | 0.94 | Accept (Poster) | | 229 | 6.67 | Dba: Distributed Backdoor Attacks Against Federated Learning | 6, 8, 6 | 0.94 | Accept (Poster) | | 230 | 6.67 | Sample Efficient Policy Gradient Methods With Recursive Variance Reduction | 6, 8, 6 | 0.94 | Accept (Poster) | | 231 | 6.67 | Efficient Transformer For Mobile Applications | 6, 8, 6 | 0.94 | Accept (Poster) | | 232 | 6.67 | Exploring Model-based Planning With Policy Networks | 6, 8, 6 | 0.94 | Accept (Poster) | | 233 | 6.67 | Multi-scale Representation Learning For Spatial Feature Distributions Using Grid Cells | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 234 | 6.67 | Variational Autoencoders For Highly Multivariate Spatial Point Processes Intensities | 6, 8, 6 | 0.94 | Accept (Poster) | | 235 | 6.67 | Can Gradient Clipping Mitigate Label Noise? | 6, 6, 8 | 0.94 | Accept (Poster) | | 236 | 6.67 | Rethinking The Security Of Skip Connections In Resnet-like Neural Networks | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 237 | 6.67 | Reinforcement Learning Based Graph-to-sequence Model For Natural Question Generation | 6, 6, 8 | 0.94 | Accept (Poster) | | 238 | 6.67 | Deep Neuroethology Of A Virtual Rodent | 6, 6, 8 | 0.94 | Accept (Spotlight) | | 239 | 6.67 | Mutual Mean-teaching: Pseudo Label Refinery For Unsupervised Domain Adaptation On Person Re-identification | 6, 8, 6 | 0.94 | Accept (Poster) | | 240 | 6.67 | Pretrained Encyclopedia: Weakly Supervised Knowledge-pretrained Language Model | 6, 6, 8 | 0.94 | Accept (Poster) | | 241 | 6.67 | Learned Step Size Quantization | 6, 6, 8 | 0.94 | Accept (Poster) | | 242 | 6.67 | Genesis: Generative Scene Inference And Sampling With Object-centric Latent Representations | 6, 6, 8 | 0.94 | Accept (Poster) | | 243 | 6.67 | Transformer-xh: Multi-hop Question Answering With Extra Hop Attention | 6, 8, 6 | 0.94 | Accept (Poster) | | 244 | 6.67 | Pc-darts: Partial Channel Connections For Memory-efficient Architecture Search | 6, 6, 8 | 0.94 | Accept (Spotlight) | | 245 | 6.67 | Neural Machine Translation With Universal Visual Representation | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 246 | 6.67 | Learning The Arrow Of Time For Problems In Reinforcement Learning | 6, 6, 8 | 0.94 | Accept (Poster) | | 247 | 6.67 | Adaptive Correlated Monte Carlo For Contextual Categorical Sequence Generation | 6, 6, 8 | 0.94 | Accept (Poster) | | 248 | 6.67 | N-beats: Neural Basis Expansion Analysis For Interpretable Time Series Forecasting | 6, 6, 8 | 0.94 | Accept (Poster) | | 249 | 6.67 | Measuring Compositional Generalization: A Comprehensive Method On Realistic Data | 6, 8, 6 | 0.94 | Accept (Poster) | | 250 | 6.67 | Pitfalls Of In-domain Uncertainty Estimation And Ensembling In Deep Learning | 6, 6, 8 | 0.94 | Accept (Poster) | | 251 | 6.67 | Tranquil Clouds: Neural Networks For Learning Temporally Coherent Features In Point Clouds | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 252 | 6.67 | Mixout: Effective Regularization To Finetune Large-scale Pretrained Language Models | 6, 8, 6 | 0.94 | Accept (Poster) | | 253 | 6.67 | Dynamically Pruned Message Passing Networks For Large-scale Knowledge Graph Reasoning | 6, 8, 6 | 0.94 | Accept (Poster) | | 254 | 6.67 | Towards Hierarchical Importance Attribution: Explaining Compositional Semantics For Neural Sequence Models | 6, 6, 8 | 0.94 | Accept (Spotlight) | | 255 | 6.67 | Real Or Not Real, That Is The Question | 8, 6, 6 | 0.94 | Accept (Spotlight) | | 256 | 6.67 | Inductive Matrix Completion Based On Graph Neural Networks | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 257 | 6.67 | The Break-even Point On The Optimization Trajectories Of Deep Neural Networks | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 258 | 6.67 | Understanding And Improving Information Transfer In Multi-task Learning | 8, 6, 6 | 0.94 | Accept (Poster) | | 259 | 6.67 | Abductive Commonsense Reasoning | 6, 8, 6 | 0.94 | Accept (Poster) | | 260 | 6.67 | Information Geometry Of Orthogonal Initializations And Training | 6, 8, 6 | 0.94 | Accept (Poster) | | 261 | 6.67 | Hoppity: Learning Graph Transformations To Detect And Fix Bugs In Programs | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 262 | 6.67 | Tree-structured Attention With Hierarchical Accumulation | 6, 6, 8 | 0.94 | Accept (Poster) | | 263 | 6.67 | Pay Attention To Features, Transfer Learn Faster Cnns | 8, 6, 6 | 0.94 | Accept (Poster) | | 264 | 6.67 | Order Learning And Its Application To Age Estimation | 6, 6, 8 | 0.94 | Accept (Poster) | | 265 | 6.67 | Gradientless Descent: High-dimensional Zeroth-order Optimization | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 266 | 6.67 | Knowledge Consistency Between Neural Networks And Beyond | 6, 8, 6 | 0.94 | Accept (Poster) | | 267 | 6.67 | Disentanglement Through Nonlinear Ica With General Incompressible-flow Networks (gin) | 8, 6, 6 | 0.94 | Accept (Spotlight) | | 268 | 6.67 | Fooling Detection Alone Is Not Enough: Adversarial Attack Against Multiple Object Tracking | 8, 6, 6 | 0.94 | Accept (Poster) | | 269 | 6.67 | Learning From Rules Generalizing Labeled Exemplars | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 270 | 6.67 | Detecting And Diagnosing Adversarial Images With Class-conditional Capsule Reconstructions | 6, 8, 6 | 0.94 | Accept (Poster) | | 271 | 6.67 | Compression Based Bound For Non-compressed Network: Unified Generalization Error Analysis Of Large Compressible Deep Neural Network | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 272 | 6.67 | Probabilistic Modeling The Hidden Layers Of Deep Neural Networks | 8, 6, 6 | 0.94 | Reject | | 273 | 6.67 | On The Geometry And Learning Low-dimensional Embeddings For Directed Graphs | 6, 6, 8 | 0.94 | Accept (Poster) | | 274 | 6.67 | Drawing Early-bird Tickets: Toward More Efficient Training Of Deep Networks | 8, 6, 6 | 0.94 | Accept (Spotlight) | | 275 | 6.67 | Variational Recurrent Models For Solving Partially Observable Control Tasks | 6, 6, 8 | 0.94 | Accept (Poster) | | 276 | 6.67 | Ensemble Distribution Distillation | 6, 6, 8 | 0.94 | Accept (Poster) | | 277 | 6.67 | Posterior Sampling For Multi-agent Reinforcement Learning: Solving Extensive Games With Imperfect Information | 6, 6, 8 | 0.94 | Accept (Talk) | | 278 | 6.67 | Improving Evolutionary Strategies With Generative Neural Networks | 6, 6, 8 | 0.94 | Reject | | 279 | 6.67 | In Search For A Sat-friendly Binarized Neural Network Architecture | 8, 6, 6 | 0.94 | Accept (Poster) | | 280 | 6.67 | Understanding The Functional And Structural Differences Across Excitatory And Inhibitory Neurons | 6, 6, 8 | 0.94 | Reject | | 281 | 6.67 | Reinforcement Learning With Competitive Ensembles Of Information-constrained Primitives | 8, 6, 6 | 0.94 | Accept (Poster) | | 282 | 6.67 | Discrepancy Ratio: Evaluating Model Performance When Even Experts Disagree On The Truth | 6, 6, 8 | 0.94 | Accept (Poster) | | 283 | 6.67 | Prediction, Consistency, Curvature: Representation Learning For Locally-linear Control | 6, 8, 6 | 0.94 | Accept (Poster) | | 284 | 6.67 | Reducing Transformer Depth On Demand With Structured Dropout | 6, 6, 8 | 0.94 | Accept (Poster) | | 285 | 6.67 | Toward Amortized Ranking-critical Training For Collaborative Filtering | 6, 6, 8 | 0.94 | Accept (Poster) | | 286 | 6.67 | Black-box Adversarial Attack With Transferable Model-based Embedding | 6, 8, 6 | 0.94 | Accept (Poster) | | 287 | 6.67 | A Neural Dirichlet Process Mixture Model For Task-free Continual Learning | 8, 6, 6 | 0.94 | Accept (Poster) | | 288 | 6.67 | Neurquri: Neural Question Requirement Inspector For Answerability Prediction In Machine Reading Comprehension | 6, 6, 8 | 0.94 | Accept (Poster) | | 289 | 6.67 | Geom-gcn: Geometric Graph Convolutional Networks | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 290 | 6.67 | Provable Robustness Against All Adversarial -perturbations For | 6, 8, 6 | 0.94 | Accept (Poster) | | 291 | 6.67 | A Probabilistic Formulation Of Unsupervised Text Style Transfer | 8, 6, 6 | 0.94 | Accept (Spotlight) | | 292 | 6.67 | A Function Space View Of Bounded Norm Infinite Width Relu Nets: The Multivariate Case | 6, 6, 8 | 0.94 | Accept (Poster) | | 293 | 6.67 | Hilloc: Lossless Image Compression With Hierarchical Latent Variable Models | 6, 6, 8 | 0.94 | Accept (Poster) | | 294 | 6.67 | Revisiting Self-training For Neural Sequence Generation | 6, 6, 8 | 0.94 | Accept (Poster) | | 295 | 6.67 | Learning Representations For Binary-classification Without Backpropagation | 8, 6, 6 | 0.94 | Accept (Poster) | | 296 | 6.67 | Model Based Reinforcement Learning For Atari | 6, 8, 6 | 0.94 | Accept (Spotlight) | | 297 | 6.67 | Ride: Rewarding Impact-driven Exploration For Procedurally-generated Environments | 6, 6, 8 | 0.94 | Accept (Poster) | | 298 | 6.67 | From Variational To Deterministic Autoencoders | 6, 8, 6 | 0.94 | Accept (Poster) | | 299 | 6.67 | Uncertainty-guided Continual Learning With Bayesian Neural Networks | 8, 6, 6 | 0.94 | Accept (Poster) | | 300 | 6.67 | Making Efficient Use Of Demonstrations To Solve Hard Exploration Problems | 6, 8, 6 | 0.94 | Accept (Poster) | | 301 | 6.67 | A Theoretical Analysis Of The Number Of Shots In Few-shot Learning | 8, 6, 6 | 0.94 | Accept (Poster) | | 302 | 6.67 | Lagrangian Fluid Simulation With Continuous Convolutions | 6, 8, 6 | 0.94 | Accept (Poster) | | 303 | 6.50 | A Closer Look At The Approximation Capabilities Of Neural Networks | 8, 6, 6, 6 | 0.87 | Accept (Poster) | | 304 | 6.50 | Dynamic Time Lag Regression: Predicting What & When | 8, 6, 6, 6 | 0.87 | Accept (Poster) | | 305 | 6.50 | Rethinking Softmax Cross-entropy Loss For Adversarial Robustness | 8, 6, 6, 6 | 0.87 | Accept (Poster) | | 306 | 6.50 | Learning Compositional Koopman Operators For Model-based Control | 6, 6, 6, 8 | 0.87 | Accept (Spotlight) | | 307 | 6.50 | Quantifying Point-prediction Uncertainty In Neural Networks Via Residual Estimation With An I/o Kernel | 6, 6, 8, 6 | 0.87 | Accept (Poster) | | 308 | 6.50 | Learning To Guide Random Search | 8, 6, 6, 6 | 0.87 | Accept (Poster) | | 309 | 6.50 | Deepv2d: Video To Depth With Differentiable Structure From Motion | 6, 6, 6, 8 | 0.87 | Accept (Poster) | | 310 | 6.33 | Coherent Gradients: An Approach To Understanding Generalization In Gradient Descent-based Optimization | 8, 8, 3 | 2.36 | Accept (Poster) | | 311 | 6.33 | Encoder-agnostic Adaptation For Conditional Language Generation | 3, 8, 8 | 2.36 | Reject | | 312 | 6.33 | Gauge Equivariant Spherical Cnns | 3, 8, 8 | 2.36 | Reject | | 313 | 6.33 | Fantastic Generalization Measures And Where To Find Them | 8, 3, 8 | 2.36 | Accept (Poster) | | 314 | 6.33 | Unsupervised Progressive Learning And The Stam Architecture | 8, 8, 3 | 2.36 | Reject | | 315 | 6.33 | Variational Template Machine For Data-to-text Generation | 8, 3, 8 | 2.36 | Accept (Poster) | | 316 | 6.33 | Automated Relational Meta-learning | 3, 8, 8 | 2.36 | Accept (Poster) | | 317 | 6.33 | Lazy-cfr: Fast And Near-optimal Regret Minimization For Extensive Games With Imperfect Information | 3, 8, 8 | 2.36 | Accept (Poster) | | 318 | 6.33 | Single Episode Transfer For Differing Environmental Dynamics In Reinforcement Learning | 3, 8, 8 | 2.36 | Accept (Poster) | | 319 | 6.33 | Transferable Perturbations Of Deep Feature Distributions | 8, 3, 8 | 2.36 | Accept (Poster) | | 320 | 6.33 | Weakly Supervised Disentanglement With Guarantees | 8, 8, 3 | 2.36 | Accept (Poster) | | 321 | 6.33 | Learning-augmented Data Stream Algorithms | 3, 8, 8 | 2.36 | Accept (Poster) | | 322 | 6.33 | Understanding Knowledge Distillation In Non-autoregressive Machine Translation