kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
PREDILECT: Preferences Delineated with Zero-Shot Language-based Reasoning in Reinforcement Learning
KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Robotics, Perception and Learning, RPL.ORCID iD: 0000-0001-5727-8140
KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Robotics, Perception and Learning, RPL.ORCID iD: 0000-0002-3510-5481
KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Robotics, Perception and Learning, RPL.ORCID iD: 0000-0002-2212-4325
2024 (English)In: HRI 2024 - Proceedings of the 2024 ACM/IEEE International Conference on Human-Robot Interaction, Association for Computing Machinery (ACM) , 2024, p. 259-268Conference paper, Published paper (Refereed)
Abstract [en]

Preference-based reinforcement learning (RL) has emerged as a new field in robot learning, where humans play a pivotal role in shaping robot behavior by expressing preferences on different sequences of state-action pairs. However, formulating realistic policies for robots demands responses from humans to an extensive array of queries. In this work, we approach the sample-efficiency challenge by expanding the information collected per query to contain both preferences and optional text prompting. To accomplish this, we leverage the zero-shot capabilities of a large language model (LLM) to reason from the text provided by humans. To accommodate the additional query information, we reformulate the reward learning objectives to contain flexible highlights - state-action pairs that contain relatively high information and are related to the features processed in a zero-shot fashion from a pretrained LLM. In both a simulated scenario and a user study, we reveal the effectiveness of our work by analyzing the feedback and its implications. Additionally, the collective feedback collected serves to train a robot on socially compliant trajectories in a simulated social navigation landscape. We provide video examples of the trained policies at https://sites.google.com/view/rl-predilect.

Place, publisher, year, edition, pages
Association for Computing Machinery (ACM) , 2024. p. 259-268
Series
ACM/IEEE International Conference on Human-Robot Interaction, ISSN 2167-2148
Keywords [en]
Human-in-the-loop Learning, Interactive learning, Preference learning, Reinforcement learning
National Category
Robotics and automation
Identifiers
URN: urn:nbn:se:kth:diva-344936DOI: 10.1145/3610977.3634970Scopus ID: 2-s2.0-85188450390OAI: oai:DiVA.org:kth-344936DiVA, id: diva2:1848562
Conference
19th Annual ACM/IEEE International Conference on Human-Robot Interaction, HRI 2024, Boulder, United States of America, Mar 11 2024 - Mar 15 2024
Note

QC 20240404

Part of ISBN 979-840070322-5

Available from: 2024-04-03 Created: 2024-04-03 Last updated: 2025-03-07Bibliographically approved
In thesis
1. Improving Sample-efficiency of Reinforcement Learning from Human Feedback
Open this publication in new window or tab >>Improving Sample-efficiency of Reinforcement Learning from Human Feedback
2025 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

With the rapid advancement of AI, the technology has moved out of the industrial and lab setting and into the hands of everyday people. Once AI and robot agents are placed in everyday households they need to be able to take into account human needs. With methods like Reinforcement Learning from Human Feedback (RLHF), the agent can learn desirable behavior by learning a reward function or optimizing a policy directly based on their feedback. Unlike vision models and large language models (LLM) which benefit from internet-scale data, RLHF is limited by the amount of feedback provided since it requires additional human effort. In this thesis, we look into how we can decrease the amount of feedback humans provide to reduce their burden when estimating a reward function without degrading the estimate. We investigate the fundamental trade-off between the informativeness and efficiency of feedback from a preference-based learning perspective. In this regard, we introduce multiple methods that can be categorized into two groups, implicit methods that increase the quality of the feedback without additional human effort, and explicit methods that aim to drastically increase the information content by using additional feedback types. To implicitly improve the efficiency of preference feedback, we look into how we can utilize Active Learning (AL) to improve the diversity of samples by strategically picking from different clusters in a learned representation through a Variational Autoencoder (VAE). Furthermore, we make use of the unique relationship between preference pairs to perform data synthesis by interpolation on the latent space of the VAE. While the implicit methods have the benefit of requiring no extra effort, they still suffer from the limited amount of information that preferences alone can provide. One limitation of preferences on trajectories is that there is no discounting which means that if a trajectory is preferred, the assumption is that the whole trajectory is preferred leading to casual confusion. Therefore, we introduce a new form of feedback called highlights that lets the user show on the trajectory, which part was good and which part was bad. Furthermore, leveraging LLMs we create a method for letting humans explain their preferences via natural language to deduce which parts were preferred. Overall, this thesis takes a step away from the assumption of internet-scale data and shows how we can achieve alignment from less human feedback.

Abstract [sv]

Med den snabba utvecklingen av AI har teknologin lämnat den industriella och laboratoriebaserade miljön och hamnat i händerna på vanliga människor. När AI- och robotagenter placeras i vardagliga hushåll måste de kunna ta hänsyn till mänskliga behov. Med metoder som Reinforcement Learning from Human Feedback (RLHF) kan en agent lära sig önskvärt beteende genom att antingen lära sig en belöningsfunktion eller optimera en policy direkt baserat på mänsklig feedback. Till skillnad från visionsmodeller och stora språkmodeller (LLM), som gynnas av internet-skaliga datamängder, är RLHF begränsat av mängden feedback som ges, eftersom det kräver ytterligare mänsklig insats.I denna avhandling undersöker vi hur man kan minska mängden feedback som människor behöver ge för att minska deras börda vid estimering av en belöningsfunktion, utan att försämra uppskattningen. Vi undersöker den fundamentala avvägningen mellan informationsinnehållet och effektiviteten i feedback från ett preferensbaserat inlärningsperspektiv. I detta avseende introducerar vi flera metoder som kan kategoriseras i två grupper: implicita metoder, som förbättrar kvaliteten på feedback utan extra mänsklig ansträngning, och explicita metoder, som syftar till att drastiskt öka informationsinnehållet genom att använda ytterligare typer av feedback.För att implicit förbättra effektiviteten av preferensfeedback undersöker vi hur Active Learning (AL) kan användas för att förbättra mångfalden av urval genom att strategiskt välja från olika kluster i en inlärd representation med hjälp av en Variational Autoencoder (VAE). Vidare utnyttjar vi den unika relationen mellan preferenspar för att utföra datasyntes genom interpolation i VAE:s latenta utrymme.Även om de implicita metoderna har fördelen att de inte kräver extra ansträngning, lider de fortfarande av den begränsade mängd information som preferenser ensamma kan ge. En begränsning med preferenser på trajektorier är att det saknas diskontering, vilket innebär att om en trajektori föredras, antas det att hela trajektorin föredras, vilket kan leda till kausal förvirring. Därför introducerar vi en ny form av feedback, kallad highlights, där användaren kan markera på trajektorier vilka delar som var bra och vilka som var dåliga. Vidare utnyttjar vi LLM:er för att skapa en metod där människor kan förklara sina preferenser genom naturligt språk för att dra slutsatser om vilka delar som föredrogs.Sammanfattningsvis tar denna avhandling ett steg bort från antagandet om internet-skaliga datamängder och visar hur vi kan uppnå anpassning med mindre mänsklig feedback.

Place, publisher, year, edition, pages
Stockholm: KTH Royal Institute of Technology, 2025. p. ix, 64
Series
TRITA-EECS-AVL ; 2025:31
Keywords
RLHF, Reinforcement Learning from Human Feedback, Reinforcement Learning, Machine Learning
National Category
Computer and Information Sciences
Research subject
Computer Science
Identifiers
urn:nbn:se:kth:diva-360983 (URN)978-91-8106-221-2 (ISBN)
Public defence
2025-04-01, https://kth-se.zoom.us/j/62755931085, F3 (Flodis), Lindstedtsvägen 26, Stockholm, 14:00 (English)
Opponent
Supervisors
Note

QC 20250307

Available from: 2025-03-07 Created: 2025-03-07 Last updated: 2025-03-17Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Holk, SimonMarta, DanielLeite, Iolanda

Search in DiVA

By author/editor
Holk, SimonMarta, DanielLeite, Iolanda
By organisation
Robotics, Perception and Learning, RPL
Robotics and automation

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 93 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf