Samuel KaskiApr 6, 2023
Modelling needs an anti-copernican revolution! But make no mistake - this is a means to making modelling more, not less principled. With @mert_celikok @MurenaPierre @FCAI_fi @idsai_uom #TuringAIFellows https://www.frontiersin.org/articles/10.3389/frai.2023.1097891/full?&utm_source=Email_to_authors_&utm_medium=Email&utm_content=T1_11.5e1_author&utm_campaign=Email_publication&field=&journalName=Frontiers_in_Artificial_Intelligence&id=1097891
Modeling needs user modeling

Modeling has actively tried to take the human out of the loop, originally for objectivity and recently also for automation. We argue that an unnecessary side effect has been that modeling workflows and machine learning pipelines have become restricted to only well-specified problems. Putting the humans back into the models would enable modeling a broader set of problems, through iterative modeling processes in which AI can offer collaborative assistance. However, this requires advances in how we scope our modeling problems, and in the user models. In this perspective article, we characterize the required user models and the challenges ahead for realizing this vision, which would enable new interactive modeling workflows, and human-centric or human-compatible machine learning pipelines.

Frontiers
Samuel KaskiMar 30, 2023
Beginning of an AI-assisted virtual laboratory in synthetic biology. More about the #VAILab mission in general: https://www.techrxiv.org/articles/preprint/Virtual_Laboratories_Transforming_research_with_AI/20412540 . In collaboration with @FCAI_fi @turinginst @OfficialUoM and several companies, large and small. #TuringAIFellows
---
RT @CSAalto
Read our story on how scientists from computer science and synthetic biology are working togethe…
https://twitter.com/CSAalto/status/1641322765381693442
Virtual Laboratories: Transforming research with AI

New scientific knowledge is needed more urgently than ever, to address global challenges such as climate change, sustainability, health and societal well-being. Could artificial intelligence (AI) accelerate the scientific process to meet global challenges in time? AI is already revolutionizing individual scientific disciplines, but we argue here that it could be more holistic and encompassing. We introduce the concept of \textit{virtual laboratories} as a new perspective on scientific knowledge generation and a means to incentivize new AI research and development. Despite the often perceived domain-specific research practices and inherent tacit knowledge, we argue that many elements of the research process generalize across  scientific domains, and that it is possible to build a common software layer that serves different domains and provides AI assistance. We outline how virtual laboratories will make it easier for AI researchers to contribute to a broad range of scientific domains, and highlight the mutual benefits virtual laboratories offer to both AI and domain scientists.

figshare
Samuel KaskiMar 6, 2023
(Engineering) design is a nice example of sequential decision making where the goals are initially unclear or evolving, and AI assistance needs new types of user models. We tell more in a perspective piece in AI Magazine, and AAAI paper @FCAI_fi #TuringAIFellows @idsai_uom
---
RT @FCAI_fi
#AI-assistance can save a designer lots of effort - so what's the bottleneck? Communicating goals & keeping humans in the loop ➡️ our researchers are dev…
https://twitter.com/FCAI_fi/status/1632732904647172097
Finnish Center for AI 🐘 @[email protected] on Twitter

“#AI-assistance can save a designer lots of effort - so what's the bottleneck? Communicating goals & keeping humans in the loop ➡️ our researchers are developing the principles & systems to make it work 🦾🤝 New from FCAI's @samikaski @oulasvirta & co.: *⃣ https://t.co/Bm99TkGkrQ”

Twitter
Show threadSamuel KaskiMar 1, 2023
With @MikkolaPetrus , Julien Martinelli, Louis Filstroff @FCAI_fi #TuringAIFellows @idsai_uom
Samuel KaskiFeb 9, 2023

The first author Sebastiaan De Peuter does not follow twitter but is certainly wotrh talking with - I am proud of this paper, on Collaborative AI for design problems and sequential decision making more generally. @FCAI_fi #TuringAIFellows @idsai_uom
---
RT @FCAI_fi
Sunday Feb. 12 at #AAAI23 in Washington: AI assistance + automation for solving sequential decision problems.

Paper: Zero-Shot Assistance in Sequential Decision Problems (@sami…
https://twitter.com/FCAI_fi/status/1623780021847355392

Finnish Center for AI 🐘 @[email protected] on Twitter

“Sunday Feb. 12 at #AAAI23 in Washington: AI assistance + automation for solving sequential decision problems. Paper: Zero-Shot Assistance in Sequential Decision Problems (@samikaski et al.) https://t.co/9W23fh4TgU”

Twitter
Samuel KaskiJan 3, 2023
Human-in-the-loop assisted de novo molecular design: Simulated proof of concept that user modelling helps the designer direct a RL-based algorithm in molecular design. Collaboration with Ola Engkvist's s team. https://doi.org/10.1186/s13321-022-00667-8 @FCAI_fi @idsai_uom #TuringAIFellows
Human-in-the-loop assisted de novo molecular design - Journal of Cheminformatics

A de novo molecular design workflow can be used together with technologies such as reinforcement learning to navigate the chemical space. A bottleneck in the workflow that remains to be solved is how to integrate human feedback in the exploration of the chemical space to optimize molecules. A human drug designer still needs to design the goal, expressed as a scoring function for the molecules that captures the designer’s implicit knowledge about the optimization task. Little support for this task exists and, consequently, a chemist usually resorts to iteratively building the objective function of multi-parameter optimization (MPO) in de novo design. We propose a principled approach to use human-in-the-loop machine learning to help the chemist to adapt the MPO scoring function to better match their goal. An advantage is that the method can learn the scoring function directly from the user’s feedback while they browse the output of the molecule generator, instead of the current manual tuning of the scoring function with trial and error. The proposed method uses a probabilistic model that captures the user’s idea and uncertainty about the scoring function, and it uses active learning to interact with the user. We present two case studies for this: In the first use-case, the parameters of an MPO are learned, and in the second use-case a non-parametric component of the scoring function to capture human domain knowledge is developed. The results show the effectiveness of the methods in two simulated example cases with an oracle, achieving significant improvement in less than 200 feedback queries, for the goals of a high QED score and identifying potent molecules for the DRD2 receptor, respectively. We further demonstrate the performance gains with a medicinal chemist interacting with the system. Graphical Abstract

BioMed Central
Show threadSamuel KaskiDec 21, 2022
Doctoral student positions in my group @AaltoUniversity , Helsinki: Topics 10-14 https://hict.fi/open-positions/ DL Jan 29 . All these with collaboration options @FCAI_fi @idsai_uom #TuringAIFellows
Helsinki ICT network: Doctoral student positions in computer science--Open positions - Helsinki Doctoral Education Network in ICT

The Helsinki Doctoral Education Network in Information and Communications Technology (HICT) is a joint initiative by Aalto University and the University of Helsinki, the two leading universities within this area in Finland. The network involves at present over 80 professors and over 200 doctoral students, and the participating units graduate altogether more than…

Helsinki Doctoral Education Network in ICT
Samuel KaskiDec 18, 2022
Thank you Anirbit Mukherjee for hosting us in the invited session on Recent Developments in Learning Theory @CFECMStatistics ! With @HeinonenMarkus we reviewed the path From differential learning to diffusion models. @FCAI_fi @idsai_uom #TuringAIFellows 1/2
Samuel KaskiDec 3, 2022
Very nice ending to the excellent #NeurIPS22 HILL Human-in-the-loop learning workshop: Best paper award to Differentiable User Models https://arxiv.org/abs/2211.16277. Congrats to Alex Hämäläinen and @mert_celikok @FCAI_fi @idsai_uom #TuringAIFellows
Differentiable User Models

Probabilistic user modeling is essential for building machine learning systems in the ubiquitous cases with humans in the loop. However, modern advanced user models, often designed as cognitive behavior simulators, are incompatible with modern machine learning pipelines and computationally prohibitive for most practical applications. We address this problem by introducing widely-applicable differentiable surrogates for bypassing this computational bottleneck; the surrogates enable computationally efficient inference with modern cognitive models. We show experimentally that modeling capabilities comparable to the only available solution, existing likelihood-free inference methods, are achievable with a computational cost suitable for online applications. Finally, we demonstrate how AI-assistants can now use cognitive models for online interaction in a menu-search task, which has so far required hours of computation during interaction.

arXiv.org
Samuel KaskiNov 28, 2022
ML action is picking up speed @OfficialUoM. This is where to meet us in #NeurIPS2022 : https://www.idsai.manchester.ac.uk/2022/11/25/manchesters-presence-at-neurips-2022/
Reach out to discuss our open positions: Lecturer/Asst Prof, RSE, Translational Research Manager, PhD students
https://www.idsai.manchester.ac.uk/research/centre-for-ai-fundamentals/news-and-opportunities/
@idsai_uom #TuringAIFellows
Manchester's presence at NeurIPS 2022 - Institute for Data Science and Artificial Intelligence

The Neural Information Processing Systems Conference 2022 is taking place in New Orleans, from 29 Nov to 9 December, 2022. A number of Manchester researchers are presenting: Human-in-the-Loop Learning (HiLL) Wei Pan, Shanghang Zhang, Pradeep Ravikumar, Vittorio Ferrari, Fisher Yu, Hao Dong, Xin Wang Multi-Mean Gaussian Processes: A novel probabilistic framework for multi-correlated longitudinal data: Arthur …

Institute for Data Science and Artificial Intelligence