Challenge encourages reproducible deep learning methods.
problem Inconsistent training behavior in deep learning competitions.
method Evaluate methods based on their training procedures, retrain in controlled settings.
result Guaranteed reproducibility and generalization of methods.
AlgoPerf competition evaluates neural network training speed-ups.
problem Improving neural network training speed using better algorithms.
method Compared 18 diverse submissions from 10 teams on multiple workloads.
result Schedule Free AdamW algorithm achieved best results in self-tuning ruleset.
Simplified image clustering achieves competitive results without text-based embeddings.
problem Complexity and resource requirements of state-of-the-art clustering methods.
method SCP: trains a small cluster head using pre-trained vision model features and positive data pairs.
result SCP achieves highly competitive performance on various benchmark datasets.
A new method splits data into independent mechanisms for better generative models.
problem Training a single model to capture the overall distribution of data.
method A competitive training procedure using mixtures of independent deep generative models and discriminators.
result The approach splits the training distribution in a sensible way and improves the quality of generated samples.
A drone catches another agile drone using competitive reinforcement learning.
problem Intercepting an agile drone with another agile drone.
method Formulated as a Competitive Reinforcement Learning problem, trained with PPO, using a high-fidelity simulation environment.
result Trained policies outperform common heuristic baselines in catch rate, time to catch, and crash rate.
This work bridges competitive learning with gradient-based learning for faster feature extraction.
problem Lack of powerful feature extractors in competitive learning methods.
method Introduces gradient-based competitive layers for feature extraction.
result Demonstrates theoretical equivalence and faster convergence of gradient-based competitive layers.
PAC-Bayes with Backprop trains neural nets with competitive error estimates and tighter risk bounds.
problem Training probabilistic neural networks with PAC-Bayes bounds.
method Two training objectives derived from PAC-Bayes bounds, evaluated on MNIST and UCI data.
result Competitive test set error estimates and tighter risk bounds than previous results.
EigenNoise provides a competitive word vector initialization scheme without pre-training data.
problem Improving word vector initialization without pre-training data.
method EigenNoise uses a dense, independent co-occurrence model to initialize word vectors.
result EigenNoise can approach GloVe performance without pre-training data.
Private cancer prediction model trained on federated genomic data.
problem Train a private cancer prediction model on federated genomic data.
method Differentially private federated learning (FL) for genomic cancer prediction.
result Ranked 3rd in a competition for private cancer prediction.
Paper tackles AI driving competition challenges with mixed simulation and real-world data.
problem AI algorithms perform poorly in real-world environments compared to simulated ones and vice versa.
method Employed imitation learning on a mixed dataset to train algorithms equally well in all environments.
result Trained algorithms performed well in both simulated and real-world environments.
New approach stabilizes GANs by leveraging implicit competitive regularization.
problem Stability issues in GAN training due to discriminator's exploitation of generator errors.
method Competitive Gradient Descent (CGD) for opponent-aware modeling of generator and discriminator.
result Significant improvement in GAN training stability without explicit regularization.
Advocacy learning uses competitive networks to improve classification accuracy.
problem Improving classification accuracy in attention-based classification problems.
method Two networks: Advocates and a Judge. Advocates produce class-conditional attention maps to convince the Judge of their class.
result Advocacy learning can lead to small improvements in classification accuracy over a baseline.
Paper assesses synthetic data quality for machine learning competitions.
problem Ensuring synthetic data maintains the performance of real data for machine learning.
method Compares synthetic data performance to real data performance for machine learning algorithms.
result Synthetic data must preserve relative performance for machine learning applications.
Transfer learning improves EEG signal classification with less data.
problem Limited data for EEG signal classification.
method Transfer learning applied to deep learning models for EEG analysis.
result Outperformed top results in BCI competition IV by 33%.
New method accelerates Pommerman training with imitation and reinforcement learning.
problem Training complex multi-agent games like Pommerman.
method Imitation learning followed by PPO reinforcement learning with modifications.
result Significantly faster training compared to other methods.
A hybrid ASR system using conformer architecture improves word-error-rate and training speed.
problem Improving word-error-rate and training efficiency for hybrid ASR systems.
method Used conformer architecture, applied time downsampling, and transposed convolutions.
result Conformer-based hybrid model achieves competitive results and significantly outperforms BLSTM-based hybrid model.
RLHC uses multiple critics at different levels to enhance RL performance.
problem Speed up RL learning and increase cumulative rewards in competition tasks.
method Introducing multiple cooperative critics at two levels of a hierarchy to provide global information.
result RLHC outperforms PPO in tennis and soccer competitions.
A simple fix helps saliency methods pass sanity checks.
problem Saliency methods fail simple sanity checks.
method Compute saliency maps for all labels, use competition to identify and remove less relevant pixels.
result CGI method is efficient, requires no additional training, and uses only the input and gradient.
A new traffic signal control method using phase competition.
problem Improving urban transportation efficiency through advanced learning techniques.
method Intuitive phase competition principle applied to reinforcement learning for traffic signal control.
result Our model achieves better solutions, faster convergence, and superior generalizability compared to existing RL methods.
Direct optimization of binary latent VAEs achieves competitive results without sampling.
problem Training VAEs with discrete latent variables using standard methods is challenging.
method Applied evolutionary algorithms to directly optimize discrete latent distributions.
result Direct optimization is efficient and competitive in zero-shot learning.
Framework for multi-agent RL with human feedback in a Snake game.
problem Improving multi-agent reinforcement learning with human feedback.
method Developed a simulated game environment for offline model training and online competitions. Introduced HILL methods and reward manipulation heuristics.
result Agents with HILL methods outperform those without in online competitions.
Combines gradient-based and competitive learning for unsupervised feature extraction.
problem Handling input data without supervision and replicating input manifold topology.
method Integrates gradient-based and competitive learning approaches to learn topological structures.
result The dual competitive layer outperforms the vanilla layer in high-dimensional datasets.
Algorithm for decentralized competition among adaptive agents.
problem Decentralized competition among adaptive networks.
method Developed an algorithm for decentralized competition among teams of adaptive agents.
result Algorithm enables decentralized competition among adaptive agents.
Competition aims to develop sample-efficient reinforcement learning methods.
problem Limited access to samples in reinforcement learning.
method Develops algorithms leveraging human demonstrations in Minecraft.
result Fosters innovation in sample-efficient reinforcement learning.
Improved reinforcement learning in Minecraft with human demonstrations.
problem Sample inefficiency in reinforcement learning.
method Training policy networks on human demonstrations first, then fine-tuning with reinforcement learning.
result Best agent achieved a mean score of 48 in Minecraft.
A game environment simulates competition among many agents for resources.
problem Understanding large-scale multiagent interactions and resource competition.
method Developed a persistent, massively multiplayer AI environment.
result Population size affects the development of skillful behaviors and niche differentiation.
New algorithm improves self-play reinforcement learning for competitive games.
problem Inefficient opponent selection in self-play reinforcement learning.
method Intelligently selects opponents based on adversarial rules derived from saddle point optimization.
result Algorithm converges to approximate equilibrium with high probability in convex-concave games.
We propose a novel learning method for multilayered neural networks which uses feedforward supervisory signal and associates classification of a new input with that of pre-trained input. The proposed method effectively uses rich input information in the earlier layer for robust leaning and revising internal representat…
Paper uses adversarial training to improve steganography.
problem Improving steganographic techniques for concealing information.
method Adversarial training applied to steganography as a discriminative task.
result Adversarial training produces a steganographic algorithm and robust steganalyzer that outperforms state-of-the-art methods.
RS-DQN protects RL agents from adversarial attacks.
problem Adversarial attacks can disrupt deep RL training and evaluation.
method Online robustness training with RS-DQN combining state-of-the-art adversarial and provably robust training.
result RS-DQN makes RL agents resilient to strong attacks.
Deep RL drone trained to compete against classical path planning in drone racing.
problem Optimizing long-term drone racing strategies using reinforcement learning.
method Used PPO algorithm on a simulated quadrotor in a racing environment created with AirSim.
result Deep RL agent outperformed classical path planning in drone racing competitions.
Ensembles of randomized decision trees, usually referred to as random forests, are widely used for classification and regression tasks in machine learning and statistics. Random forests achieve competitive predictive performance and are computationally efficient to train and test, making them excellent candidates for r…
Generative model generates realistic text without reinforcement learning.
problem Training GANs for natural language processing is challenging due to discrete sequences.
method Autoencoder learns low-dimensional sentence representation, GAN generates vectors in this space.
result Model generates realistic text with competitive BLEU scores and human ratings.
Paper studies competitive networks where teams aim to minimize their own objectives, adapting to each other's strategies.
problem Competitive networks where teams have conflicting objectives.
method Proposes diffusion learning algorithms for two classes of network games: zero-sum and non-zero-sum.
result Stability performance of proposed algorithms analyzed and demonstrated through experiments.
A new method compares image classifiers using adaptive sampling of natural images.
problem Evaluation of image classifiers on small, fixed test sets may not generalize to real-world images.
method Adaptive sampling from a large corpus of unlabeled images to maximize classifier discrepancies measured by WordNet hierarchy.
result Human labeling of model-dependent image sets reveals relative classifier performance.
Dynamic learning rate set using Mutual Information for neural networks.
problem Optimizing learning rate in deep neural networks.
method Mutual Information is used to dynamically adjust the learning rate during training cycles. Two approaches are tested: relative change and relative to a reference measure.
result Mutual Information can effectively set dynamic learning rates, leading to competitive performance.
New method uses nearest neighbors quantile filter for probabilistic energy forecasting.
problem Creating accurate probabilistic energy forecasts using complex data mining techniques.
method Uses a new nearest neighbors quantile filter to create quantile regressions without a non-differentiable cost function.
result Demonstrates superior performance in Global Energy Forecasting Competition 2014.
This paper proposes a method to learn labels from unlabeled data for better anomaly detection.
problem Learning labels from unlabeled data for better anomaly detection.
method End-to-end architecture with one encoder and two decoders trained to model inliers and outliers' data distributions competitively.
result Empirical studies show superior performance compared to state-of-the-art methods.
Before the operation of a motor imagery based brain-computer interface (BCI) adopting machine learning techniques, a cumbersome training procedure is unavoidable. The development of a practical BCI posed the challenge of classifying single-trial EEG with a small training set. In this letter, we addressed this problem b…
Study shows competition feedback can make ML predictors biased towards specific user groups.
problem How competition affects machine learning predictors and user prediction quality.
method Flexible model of competing ML predictors, empirical and mathematical analysis.
result Competition causes predictors to specialize for specific sub-populations at the cost of general performance.
Syllable-aware models perform similarly to character-based ones but use fewer parameters and train faster.
problem Improving word-level language modeling performance with syllable-based models.
method Used a syllable-aware neural language model with fewer parameters and faster training.
result Achieved comparable performance to character-based models but with 18%-33% fewer parameters and 1.2-2.2 times faster training.
Generative adversarial networks improve brain-computer interface performance with limited data.
problem Limited training samples in brain-computer interfaces.
method Conditional Deep Convolutional Generative Adversarial Networks (cDCGAN) for data augmentation.
result Generated artificial EEG data improves classification accuracy in brain-computer interface tasks.
This paper uses NARX neural networks for macroeconomic forecasting and goal setting.
problem Improving accuracy in macroeconomic forecasting and goal setting.
method Literature review and construction of specific NARX neural networks for macroeconomic indicators.
result NARX neural networks can be trained to make accurate predictions for macroeconomic indicators and national goals.
Study improves GFM accuracy with time series augmentation.
problem Limited time series data hinders GFM performance.
method Data augmentation techniques (GRATIS, MBB, DBA) and transfer learning.
result Significant improvement in GFM accuracy over baseline.
System constructs public competitor graph from financial reports.
problem Time-consuming and expert-laden manual extraction of corporate relationships.
method Financial report processing to generate reliable knowledge graph of corporate relationships.
result More than 83% of S\&P 500 companies' competition relationships retrieved.
A multi-task model tackles citation purpose classification with limited data.
problem Classifying citations based on their purpose is challenging due to limited labeled data and subjectivity.
method Combines linguistic features, TF-IDF, and an LSTM-with-attention model for multi-task learning.
result Improves classification accuracy compared to single-task models.
MGD improves ML training efficiency and accuracy.
problem Optimizing large-scale ML model training configurations.
method Metagradient descent and smooth model training.
result Significantly improved model performance and learning rate schedules.
Deep learning detects bid-rigging cartels with high accuracy.
problem Detecting bid-rigging cartels using pairwise bidding interactions.
method Convolutional neural networks applied to graphs of normalized bid values.
result Convolutional neural networks achieve around 90% accuracy in classifying collusive and competitive bidding interactions.