A Minimaximalist Approach to Reinforcement Learning from Human Feedback

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Swamy, Gokul, Dann, Christoph, Kidambi, Rahul, Wu, Zhiwei Steven, Agarwal, Alekh
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Inverse Reinforcement Learning without Reinforcement Learning
di: Swamy, Gokul, et al.
Pubblicazione: (2023)

Design Considerations in Offline Preference-based RL
di: Agarwal, Alekh, et al.
Pubblicazione: (2025)

The Virtues of Pessimism in Inverse Reinforcement Learning
di: Wu, David, et al.
Pubblicazione: (2024)

Mitigating Preference Hacking in Policy Optimization with Pessimism
di: Gupta, Dhawal, et al.
Pubblicazione: (2025)

Hybrid Inverse Reinforcement Learning
di: Ren, Juntao, et al.
Pubblicazione: (2024)

Preserving Expert-Level Privacy in Offline Reinforcement Learning
di: Sharma, Navodita, et al.
Pubblicazione: (2024)

All Roads Lead to Likelihood: The Value of Reinforcement Learning in Fine-Tuning
di: Swamy, Gokul, et al.
Pubblicazione: (2025)

Multi-Agent Imitation Learning: Value is Easy, Regret is Hard
di: Tang, Jingwu, et al.
Pubblicazione: (2024)

Non-Linear Reinforcement Learning in Large Action Spaces: Structural Conditions and Sample-efficiency of Posterior Sampling
di: Agarwal, Alekh, et al.
Pubblicazione: (2022)

Gained in Translation: Privileged Pairwise Judges Enhance Multilingual Reasoning
di: Sutawika, Lintang, et al.
Pubblicazione: (2026)

A Model Selection Approach for Corruption Robust Reinforcement Learning
di: Wei, Chen-Yu, et al.
Pubblicazione: (2021)

Back to Blackwell: Closing the Loop on Intransitivity in Multi-Objective Preference Fine-Tuning
di: Zhang, Jiahao, et al.
Pubblicazione: (2026)

Provable Multi-Party Reinforcement Learning with Diverse Human Feedback
di: Zhong, Huiying, et al.
Pubblicazione: (2024)

More Benefits of Being Distributional: Second-Order Bounds for Reinforcement Learning
di: Wang, Kaiwen, et al.
Pubblicazione: (2024)

Diffusing States and Matching Scores: A New Framework for Imitation Learning
di: Wu, Runzhe, et al.
Pubblicazione: (2024)

Offline Imitation Learning from Multiple Baselines with Applications to Compiler Optimization
di: Marinov, Teodor V., et al.
Pubblicazione: (2024)

A Human-Centric Approach to Explainable AI for Personalized Education
di: Swamy, Vinitra
Pubblicazione: (2025)

From Foresight to Forethought: VLM-In-the-Loop Policy Steering via Latent Alignment
di: Wu, Yilin, et al.
Pubblicazione: (2025)

Reinforcement Learning from Human Feedback
di: Lambert, Nathan
Pubblicazione: (2025)

Theoretical Foundations and Effective Algorithms for Policy-Aware Simulator Learning
di: Dann, Christoph, et al.
Pubblicazione: (2026)

Catoni Contextual Bandits are Robust to Heavy-tailed Rewards
di: Ye, Chenlu, et al.
Pubblicazione: (2025)

Strategyproof Reinforcement Learning from Human Feedback
di: Buening, Thomas Kleine, et al.
Pubblicazione: (2025)

Differentially Private SGD Without Clipping Bias: An Error-Feedback Approach
di: Zhang, Xinwei, et al.
Pubblicazione: (2023)

Conditional Language Policy: A General Framework for Steerable Multi-Objective Finetuning
di: Wang, Kaiwen, et al.
Pubblicazione: (2024)

Reinforcement Learning from Human Feedback: A Statistical Perspective
di: Liu, Pangpang, et al.
Pubblicazione: (2026)

Robust Reinforcement Learning from Corrupted Human Feedback
di: Bukharin, Alexander, et al.
Pubblicazione: (2024)

Dual Active Learning for Reinforcement Learning from Human Feedback
di: Liu, Pangpang, et al.
Pubblicazione: (2024)

Your Learned Constraint is Secretly a Backward Reachable Tube
di: Qadri, Mohamad, et al.
Pubblicazione: (2025)

Optimizing Pre-Training Data Mixtures with Mixtures of Data Expert Models
di: Belenki, Lior, et al.
Pubblicazione: (2025)

REBEL: Reward Regularization-Based Approach for Robotic Reinforcement Learning from Human Feedback
di: Chakraborty, Souradip, et al.
Pubblicazione: (2023)

Exploring Data Scaling Trends and Effects in Reinforcement Learning from Human Feedback
di: Shen, Wei, et al.
Pubblicazione: (2025)

Dense Reward for Free in Reinforcement Learning from Human Feedback
di: Chan, Alex J., et al.
Pubblicazione: (2024)

Multi-turn Reinforcement Learning from Preference Human Feedback
di: Shani, Lior, et al.
Pubblicazione: (2024)

Reinforcement Learning from Multi-level and Episodic Human Feedback
di: Elahi, Muhammad Qasim, et al.
Pubblicazione: (2025)

A Survey of Reinforcement Learning from Human Feedback
di: Kaufmann, Timo, et al.
Pubblicazione: (2023)

Efficient Imitation under Misspecification
di: Espinosa-Dice, Nicolas, et al.
Pubblicazione: (2025)

The Power of Active Multi-Task Learning in Reinforcement Learning from Human Feedback
di: Chen, Ruitao, et al.
Pubblicazione: (2024)

Low-Rank Contextual Reinforcement Learning from Heterogeneous Human Feedback
di: Lee, Seong Jin, et al.
Pubblicazione: (2024)

Data-dependent Exploration for Online Reinforcement Learning from Human Feedback
di: Zhang, Zhen-Yu, et al.
Pubblicazione: (2026)

The Alignment Ceiling: Objective Mismatch in Reinforcement Learning from Human Feedback
di: Lambert, Nathan, et al.
Pubblicazione: (2023)

if (filemtime($filename) + $maxlifetime < time()) { unlink($filename); $count++; } } return $count; } /** * A function that is called internally when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ protected function saveSession($sessId, $data): bool { $sessFile = $this->path . '/sess_' . $sessId; if ($handle = fopen($sessFile, 'w')) { $return = false; // Lock the file for exclusive access to avoid issues with multiple // processes writing session simultaneously: if (flock($handle, LOCK_EX)) { $return = fwrite($handle, $data); // Make sure that there's no trailing data by truncating the file to // the correct length: ftruncate($handle, strlen($data)); fflush($handle); flock($handle, LOCK_UN); } fclose($handle); if ($return !== false) { return true; } } // If we got this far, something went wrong with the file output... // It is tempting to throw an exception here, but this code is called // outside of the context of exception handling, so all we can do is // echo a message.

// Anecdotal testing Today and Yesterday seems to indicate destroy() // is called by the garbage collector and everything is good. // Something to keep in mind though. return 0; } /** * Write function that is called when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ public function write($sessId, $data): bool { if ($this->writesDisabled) { return true; } return $this->saveSession($sessId, $data); } /** * A function that is called internally when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ abstract protected function saveSession($sessId, $data): bool; }

* * @return void */ public function writeClose() { // The assumption is that we're using PHP's ext/session. // session_write_close() will actually overwrite $_SESSION with an // empty array on completion -- which leads to a mismatch between what // is in the storage object and $_SESSION. To get around this, we // temporarily reset $_SESSION to an array, and then re-link it to // the storage object. // // Additionally, while you _can_ write to $_SESSION following a // session_write_close() operation, no changes made to it will be // flushed to the session handler. As such, we now mark the storage // object isImmutable. $storage = $this->getStorage(); if (! $storage->isImmutable()) { $_SESSION = $storage->toArray(true); session_write_close(); $storage->fromArray($_SESSION); $storage->markImmutable(); } } /** * Attempt to set the session name * * If the session has already been started, or if the name provided fails * validation, an exception will be raised. * * @param string $name * @return SessionManager * @throws Exception\InvalidArgumentException */ public function setName($name) { if ($this->sessionExists()) { throw new Exception\InvalidArgumentException( 'Cannot set session name after a session has already started'

Environment & details:

GET Data empty

POST Data empty

Files empty

Key	Value
VUFIND_SESSION	"6131a2af78394de7b92080b537c5950f"
language	"it"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1791288579.2767 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "6131a2af78394de7b92080b537c5950f" ]`]
SessionState	Laminas\Stdlib\ArrayObject {#938}
FlashMessenger	Laminas\Stdlib\ArrayObject {#1433}
SessionHelper	Laminas\Stdlib\ArrayObject {#1759}

Server/Request Data

Key	Value
REDIRECT_VUFIND_LOCAL_DIR	"/usr/local/vufind/local"
REDIRECT_HTTPS	"on"
REDIRECT_SSL_TLS_SNI	"info.umar.mx"
REDIRECT_STATUS	"200"
VUFIND_LOCAL_DIR	"/usr/local/vufind/local"
HTTPS	"on"
SSL_TLS_SNI	"info.umar.mx"
HTTP_ACCEPT	"/"
HTTP_USER_AGENT	"Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; ClaudeBot/1.0; +claudebot@anthropic.com)"
HTTP_COOKIE	"VUFIND_SESSION=6131a2af78394de7b92080b537c5950f; language=it; ui=standard"
HTTP_ACCEPT_ENCODING	"gzip, br, zstd, deflate"
HTTP_HOST	"info.umar.mx"
PATH	"/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/snap/bin"
SERVER_SIGNATURE	"<address>Apache/2.4.66 (Ubuntu) Server at info.umar.mx Port 443</address>\n"
SERVER_SOFTWARE	"Apache/2.4.66 (Ubuntu)"
SERVER_NAME	"info.umar.mx"
SERVER_ADDR	"200.23.222.19"
SERVER_PORT	"443"
REMOTE_ADDR	"216.73.216.28"
DOCUMENT_ROOT	"/usr/local/vufind/public"
REQUEST_SCHEME	"https"
CONTEXT_PREFIX	""
CONTEXT_DOCUMENT_ROOT	"/usr/local/vufind/public"
SERVER_ADMIN	"admin@localhost"
SCRIPT_FILENAME	"/usr/local/vufind/public/index.php"
REMOTE_PORT	"44637"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2401_04056/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2401_04056/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1791288579.264
REQUEST_TIME	1791288579

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler