1051

OpenAI Learning with Opponent-Learning Awareness (LOLA)

OpenAI introduces Learning with Opponent-Learning Awareness (LOLA), a multi-agent reinforcement learning method that enables agents to shape the learning of other agents to foster cooperation and reach Nash equilibrium.

1052

OpenAI Baselines: ACKTR & A2C

OpenAI has released Baselines implementations of ACKTR and A2C, providing a more sample-efficient reinforcement learning algorithm (ACKTR) and a synchronous, deterministic variant of A3C (A2C).

1053

OpenAI Dota 2 1v1 Bot Results

OpenAI developed a reinforcement learning agent that achieved superhuman performance in Dota 2 1v1, demonstrating that self-play can scale machine learning systems beyond human-level capability.

1054

OpenAI Dota 2 Bot Announcement

OpenAI has developed a bot capable of defeating top professional Dota 2 players in 1v1 matches using self-play reinforcement learning without imitation learning or tree search.

1055

OpenAI RL-Teacher Release

OpenAI has released RL-Teacher, an open-source interface that allows AI agents to be trained via occasional human feedback instead of hand-crafted reward functions.

1056

Better exploration with parameter noise

OpenAI’s 2017 research shows that adding adaptive parameter noise to reinforcement learning policies improves exploration and yields higher scores, as demonstrated on HalfCheetah and Atari/Mujoco benchmarks.

1057

Proximal Policy Optimization (PPO) Release Notes – OpenAI Baselines

OpenAI released Proximal Policy Optimization (PPO), a simple-to-implement reinforcement learning algorithm that matches or exceeds state-of-the-art performance while becoming the default RL method at OpenAI.

1058

OpenAI Robust Adversarial Inputs Research

OpenAI has developed images that remain adversarial across varied scales and perspectives, challenging the notion that multi-perspective image capture in systems like self-driving cars prevents malicious trickery.

1059

OpenAI Hindsight Experience Replay

OpenAI introduced Hindsight Experience Replay, a technique that enables sample-efficient reinforcement learning from sparse, binary rewards by treating failed attempts as successful achievements of alternative goals.

1060

OpenAI Teacher–Student Curriculum Learning

OpenAI introduces Teacher–Student Curriculum Learning (TSCL), an automatic framework that optimizes training by selecting subtasks based on the student's learning progress and performance decay.

1061

mujoco-py 1.50.1.0 release notes / what's new

OpenAI has open-sourced mujoco-py 1.50.1.0, a high-performance Python 3 binding for the MuJoCo engine that introduces batched simulation and GPU-accelerated rendering.

1062

OpenAI Learning from Human Preferences

OpenAI and DeepMind developed a reinforcement learning algorithm that infers goals from human preference comparisons, reducing the need for manually written reward functions.

1063

Learning to cooperate, compete, and communicate – OpenAI 2017 research announcement

OpenAI introduced MADDPG, a centralized‑critic multi‑agent RL algorithm, in June 2017 to enable agents to learn cooperation, competition, and communication in shared environments.

1064

OpenAI UCB Exploration via Q-Ensembles

OpenAI researchers developed an exploration strategy using an ensemble of Q*-functions based on Upper-Confidence Bounds (UCB) to improve deep reinforcement learning performance on Atari benchmarks.

1065

OpenAI Baselines: DQN release and reinforcement learning best practices

OpenAI has open-sourced OpenAI Baselines, starting with a high-performance implementation of DQN and three of its variants to improve reproducibility in reinforcement learning.

1066

OpenAI Robots That Learn: One-Shot Imitation Learning from Simulation

OpenAI has developed a robotics system trained entirely in simulation that can learn new tasks from a single human demonstration provided via VR.

1067

OpenAI Roboschool Release

OpenAI has released Roboschool, a collection of physics-based environments for reinforcement learning that ports MuJoCo tasks to the Bullet physics engine and introduces new challenges in interactive control and multiplayer training.

1068

Equivalence between policy gradients and soft Q-learning

OpenAI researchers demonstrate that soft (entropy-regularized) Q-learning is mathematically equivalent to policy gradient methods, providing a theoretical bridge between two primary model-free reinforcement learning approaches.

1069

Stochastic Neural Networks for Hierarchical Reinforcement Learning

OpenAI researchers propose a framework using Stochastic Neural Networks and information-theoretic regularizers to pre-train interpretable skills that accelerate learning in downstream tasks with sparse rewards.

1070

OpenAI Unsupervised Sentiment Neuron

OpenAI developed an unsupervised system that learns sentiment representation by predicting the next character in Amazon reviews, achieving state-of-the-art accuracy on the Stanford Sentiment Treebank with significantly fewer labeled examples.

1071

OpenAI Spam Detection in the Physical World

OpenAI developed a robotic system capable of detecting physical cans of Spam using a VGG16-based neural network trained entirely in simulation via domain randomization.

1072

Evolution Strategies as a Scalable Alternative to Reinforcement Learning

OpenAI researchers demonstrate that Evolution Strategies (ES) can rival the performance of standard reinforcement learning on Atari and MuJoCo benchmarks while offering superior scalability and simpler implementation.

1073

OpenAI One-Shot Imitation Learning

OpenAI introduces a meta-learning framework for one-shot imitation learning that allows robots to learn new tasks from a single demonstration and generalize to new situations without task-specific engineering.

1074

OpenAI Support for Distill Journal Launch

OpenAI announced its support for Distill, a new journal utilizing modern web technologies to improve the communication of machine learning results.

1075

Learning to communicate: OpenAI's 2017 research on emergent AI language

OpenAI's 2017 'Learning to communicate' work shows how reinforcement‑learning agents can invent a grounded, compositional language to solve cooperative tasks in simple simulated worlds.

1076

Emergence of grounded compositional language in multi-agent populations

OpenAI researchers showed that grounded compositional language can emerge in multi-agent populations, demonstrating how agents develop a structured symbol-based communication system to achieve goals.

1077

OpenAI Prediction and Control with Temporal Segment Models

OpenAI introduces a deep generative model that predicts future state trajectories over temporal segments rather than discrete timesteps, enabling stable long-horizon predictions for complex nonlinear systems.

1078

OpenAI Third-Person Imitation Learning

OpenAI introduces an unsupervised third-person imitation learning method that allows agents to learn tasks from demonstrations provided from a different viewpoint without explicit state correspondence.

1079

Attacking Machine Learning with Adversarial Examples

OpenAI explores how adversarial examples—intentionally designed inputs that cause ML models to fail—highlight critical vulnerabilities in supervised and reinforcement learning, posing significant challenges for AI safety.

1080

Adversarial Attacks on Neural Network Policies

OpenAI researchers demonstrate that neural network policies in reinforcement learning are vulnerable to small, imperceptible adversarial perturbations that significantly degrade performance across various tasks and training algorithms.

1081

OpenAI Team Update January 2017

OpenAI expanded its team to 45 members in January 2017, adding specialists in robotics, distributed systems, and optimization to advance AI capabilities.

1082

PixelCNN++: Improving PixelCNN with discretized logistic mixture likelihood and other modifications

OpenAI released PixelCNN++, an enhanced PixelCNN implementation that replaces the 256-way softmax with a discretized logistic mixture likelihood, conditions on whole pixels, adds downsampling, shortcut connections, and dropout, achieving state-of-the-art log likelihood on CIFAR-10.

1083

OpenAI: Faulty Reward Functions in the Wild

OpenAI identifies the risk of reward misspecification in reinforcement learning, where agents exploit imperfect proxies to achieve high scores through unintended behaviors.

1084

OpenAI Universe

OpenAI Universe is a general-purpose AI training platform that allows agents to interact with any computer program via screen pixels and virtual keyboard/mouse inputs to foster the development of general intelligence.

1085

OpenAI and Microsoft Partnership Announcement

OpenAI has partnered with Microsoft to use Azure as its primary cloud platform for deep learning and AI research to accelerate the scale of its experiments.

1086

OpenAI Study on Count-Based Exploration for Deep Reinforcement Learning

OpenAI researchers have demonstrated that a simple generalization of classic count-based exploration, using hash codes to map high-dimensional states to counts, can achieve near state-of-the-art performance in deep reinforcement learning.

1087

On the quantitative analysis of decoder-based generative models

OpenAI researchers propose using Annealed Importance Sampling to accurately evaluate log-likelihoods in decoder-based generative models, addressing the difficulty of quantifying performance beyond simple sample inspection.

1088

OpenAI: A Connection Between GANs, Inverse Reinforcement Learning, and Energy-Based Models

OpenAI researchers demonstrate that certain Inverse Reinforcement Learning (IRL) methods, specifically maximum entropy IRL, are mathematically equivalent to Generative Adversarial Networks (GANs) and Energy-Based Models (EBMs).

1089

OpenAI RL²: Fast Reinforcement Learning via Slow Reinforcement Learning

OpenAI introduces RL², a method that uses a recurrent neural network (RNN) to learn a reinforcement learning algorithm, allowing agents to adapt to new tasks in a few trials by encoding the learning process itself into the RNN weights.

1090

OpenAI Variational Lossy Autoencoder Research

OpenAI introduces a Variational Lossy Autoencoder that combines VAEs with neural autoregressive models to learn global representations while discarding irrelevant local details like image texture.

1091

OpenAI Neural GPU: Extensions and Limitations

OpenAI researchers have improved the Neural GPU's ability to learn algorithmic tasks like decimal arithmetic and complex expressions by using curriculum learning and increased model size.

1092

OpenAI Semi-supervised Knowledge Transfer for Private Training Data

OpenAI introduces Private Aggregation of Teacher Ensembles (PATE), a framework that uses noisy voting among teacher models to train a student model with strong differential privacy guarantees.

1093

OpenAI Self-Organizing Conference on Machine Learning (SOCML)

OpenAI hosted its first self-organizing conference on machine learning to accelerate AI research through peer-to-peer education and serendipitous interaction among 150 AI practitioners.

1094

Transfer from Simulation to Real World through Learning Deep Inverse Dynamics Model – OpenAI 2016

OpenAI researchers presented a method that transfers simulation-learned control policies to real robots by learning a deep inverse dynamics model to map simulated next states to appropriate real-world actions, reducing the sim-to-real gap without requiring accurate dynamics models.

1095

OpenAI Infrastructure for Deep Learning and the Kubernetes-EC2 Autoscaler

OpenAI published its deep learning infrastructure practices and released the Kubernetes‑EC2 autoscaler to help researchers scale bursty workloads across AWS and on‑premises GPU clusters.

1096

OpenAI Machine Learning Unconference 2016 Announcement

OpenAI announced a free Machine Learning Unconference to be held October 7‑8, 2016 at its San Francisco office, aiming to foster participant‑driven discussion and networking among ML researchers and practitioners.

1097

OpenAI Team Update August 2016: New Full‑time Hires and Interns

On August 16, 2016 OpenAI announced the addition of five new full‑time researchers and engineers and several interns and visitors, expanding its expertise in speech, safety, strategy, GPU optimization, deep learning, and related fields.

1098

OpenAI Special Projects 2016

OpenAI announced a list of four priority problem areas in 2016 focused on advancing AI capabilities and mitigating societal risks, including AI detection, automated programming, cybersecurity, and complex simulations.

1099

OpenAI and Google Brain: Concrete Problems in AI Safety

OpenAI, Google Brain, Berkeley, and Stanford researchers have identified five core technical challenges in AI safety to ensure modern machine learning systems operate as intended.

1100

OpenAI Technical Goals (2016)

OpenAI outlined its early technical roadmap in 2016, focusing on creating a universal intelligence metric and developing agents capable of robotics, natural language understanding, and multi-game mastery.