IA2 uses Deep Reinforcement Learning to slash database runtimes by 40%, while new Hyperbolic SVM techniques utilize semidefinite relaxation. https://hackernoon.com/ai-driven-database-tuning-faster-index-selection-with-ia2-and-td3-td-swar #deepreinforcementlearning

IA2 revolutionizes index selection with rapid training, reducing SQL runtime by 61% via adaptive action pruning and workload modeling https://hackernoon.com/adaptive-action-pruning-scaling-index-selection-for-unseen-workloads #deepreinforcementlearning

IA2 uses a two-phase framework to generate states and action pools from workloads, enabling RL agents to make sequential index selection decisions. https://hackernoon.com/unseen-workload-optimization-the-two-phase-ia2-approach #deepreinforcementlearning

The TD3-TD-SWAR model advances database optimization by framing index selection as a DRL problem with adaptive action masking for faster training. https://hackernoon.com/adaptive-action-masking-accelerating-decision-making-in-database-tuning #deepreinforcementlearning

This research validates a weekly re-trained DRL agent, showing it outperforms static models & Black-Scholes for practical American option hedging. https://hackernoon.com/validating-hyperparameters-and-a-weekly-re-training-strategy-for-drl-option-hedging #deepreinforcementlearning

This methodology details how to train and test DRL agents for American option hedging, introducing a novel weekly re-training strategy using Chebyshev pricing. https://hackernoon.com/dont-just-train-your-ai-re-train-it-the-weekly-workout-plan-for-a-smarter-option-hedge #deepreinforcementlearning

This review of DRL hedging literature highlights the need for hyperparameter analysis, especially for real-world American option applications. https://hackernoon.com/avoiding-the-pitfalls-a-guide-to-the-current-state-of-drl-option-hedging-research #deepreinforcementlearning

This paper makes Deep Reinforcement Learning practical for hedging American options by optimizing hyperparameters and using a weekly re-training strategy. https://hackernoon.com/how-weekly-ai-training-is-beating-a-nobel-prize-winning-formula #deepreinforcementlearning

RoboPianist: Dexterous Piano Playing with Deep Reinforcement Learning (2023) — https://kzakka.com/robopianist/#demo
#HackerNews #RoboPianist #DeepReinforcementLearning #PianoAI #MachineLearning #Robotics #2023

Autonomy Talks - Georgia Chalvatzaki: Shaping #Robotic Assistance through Structured #Robot #Learning: https://www.youtube.com/watch?v=e0aQC3C8P7w #robotics #machinelearning

Around 12:30 they present the training of a model-free #MDP #deepreinforcementlearning using a model-based #ai #planner #aiplanner. Indeed it drastically boosts the training.

The general idea is to guide an implicit model using a model-based approximation, and it works also for assembly tasks, computer vision, pick and place…

Last but not least, came Tekgul & Asokan's "FLARE: Fingerprinting Deep Reinforcement Learning Agents using Universal Adversarial Masks" which is robust to model modification attacks. (https://www.acsac.org/2023/program/final/s264.html) 4/4
#MachineLearningSecurity #DeepReinforcementLearning #SecurityInAI

Mini-Quadkopter lernt Fliegen in Sekunden | heise online
https://heise.de/-9623443 #DeepReinforcementLearning #ReinforcementLearning #RL

📬 KI-Roboter: Meister des Labyrinth-Geschicklichkeitsspiels
#KünstlicheIntelligenz #Cheat #Cheater #Cheats #CyberRunner #DeepReinforcementLearning #github #KIRoboter #Labyrinth #ProfDrRaffaelloD’Andrea #Robotik https://tarnkappe.info/artikel/kuenstliche-intelligenz/ki-roboter-meister-des-labyrinth-geschicklichkeitsspiels-285483.html

Will the next generation of #LLM come from #DeepMind?
https://www.wired.com/story/google-deepmind-demis-hassabis-chatgpt/
They may have a shot at it given their expertise in #DeepReinforcementLearning. If their #AI can plan tasks with solid logical grounds, can't they also produce solid explanations?

An innovative bimanual robot displays tactile sensitivity close to human-level dexterity using AI https://innovationtoronto.com/2023/08/an-innovative-bimanual-robot-displays-tactile-sensitivity-close-to-human-level-dexterity-using-ai/

#BristolUni #bimanualrobot #DeepReinforcementLearning #BiTouchsystem

El lado del mal - AlphaDev: La IA que optimiza la implementación de los algoritmos mejor que los humanos https://www.elladodelmal.com/2023/06/alphadev-la-ia-que-optimiza-la.html #AlphaDev #IA #AI #Programación #Algorítmica #Algorithms #InteligenciaArtificial #DeepReinforcementLearning #Intel #ensamblador #DeepLearning #LLM

Faster sorting #algorithms discovered using #DeepReinforcementLearning | #Nature

"Here we show how #ArtificialIntelligence can go beyond the current state of the art by discovering hitherto unknown routines. To realize this, we formulated the task of finding a better sorting routine as a single-player game. We then trained a new deep #ReinforcementLearning agent, #AlphaDev, to play this game. AlphaDev discovered small sorting algorithms from scratch that outperformed previously known human benchmarks. These algorithms have been integrated into the #LLVM standard C++ sort library3. This change to this part of the sort library represents the replacement of a component with an algorithm that has been automatically discovered using reinforcement learning."

https://www.nature.com/articles/s41586-023-06004-9

#introducton 
Here to talk #neuroscience mostly about human #decisionmaking and #learning in #ReinforcementLearning environments. 
I'm doing my #PhD at LMU #Munich, working with #EyeTracking and #EEG data, using #Python and #MATLAB, and am interested in #DeepReinforcementLearning, #senseofagency and #socialagency ! 
On the side, I play #chess and hike #Alpine !
Hello #Fediverse :)