Kevin McKee
@empiricallykev
Followers
270
Following
50
Media
0
Statuses
18
Social cognition @deepmind. He/him 💖💜💙. Not myself by myself
Joined February 2021
In forthcoming work in PNAS (with @weidingerlaura, @empiricallykev, @saffronhuang and others) we look at preference elicitation in relation to principle choice for AI assistants from behind a Veil of Ignorance. This process is designed to be fair to all parties. Stay tuned!
0
3
16
Opinion piece with @kormilitzin @weballergy @empiricallykev We talk about ▪️ the double-bind of stigmatized health conditions and sensitive characteristics (1/3)
A participatory initiative to include LGBT+ voices in AI for mental health. Correspondence from @kormilitzin @weballergy @empiricallykev and @dan_w_joyce, @UniofOxford and @DeepMind
https://t.co/KjMx3P56Sk
1
4
3
A participatory initiative to include LGBT+ voices in AI for mental health. Correspondence from @kormilitzin @weballergy @empiricallykev and @dan_w_joyce, @UniofOxford and @DeepMind
https://t.co/KjMx3P56Sk
0
11
17
3 days to go ... Friday 9th December 2022 Participatory Approaches to AI for Mental Health https://t.co/gt9LhyKygS NeurIPS 2022 Virtual Workshop https://t.co/JcErjXmdgf
@kormilitzin
@weballergy
@empiricallykev
@dan_w_joyce
sites.google.com
Overview The advances in machine learning (ML) and artificial intelligence (AI) towards improving and streamlining healthcare have been widely recognised and discussed, and there is a long-standing...
0
2
1
The Cooperation Colloquia are back this week! Join us on Friday for Kevin McKee @empiricallykev (DeepMind) Social Value Orientation: From Human Cooperation to Cooperative AI November 4, 15:00 UTC+1 Sign-up here: https://t.co/xPgLbVMtxM
0
8
9
The Cooperation Colloquia are back! Nov 4: Kevin McKee / @empiricallykev Nov 17: Alessandra Cassar / @Cassar_Ale Dec 2: Delia Baldassari Dec 16: Christian Hilbe / @chilbe3 Sign up and join us for the talks: https://t.co/xPgLbVN1nk
1
10
37
If you have background at the intersection of (mental) health and AI research and would like to join the Program Committee of our upcoming PAI4MH NeurIPS workshop - we are welcoming additional volunteers to help review the paper submissions. Please DM if interested.
For those working on applications of AI to wellbeing and mental health - a reminder that the abstract deadline for the PAI4MH NeurIPS workshop is on the 25th, and the submission deadline on the 29th of September.
3
4
10
Excited to hear @farbandish, @shakir_za, @willie_agnew, @struthious, & other brilliant folks discuss the relationships—current, imminent, or possible—between queer communities and AI. It's been a blast organizing this. Please come join if you're interested!
DeepMind is hosting a workshop centered on queer community and AI on 6 July. Queer in AI members and allies are invited to join online or in-person in London. You can sign up through this form! https://t.co/BGzsysJzKO
0
4
15
For people, identity is forever in flux. To a machine, human identity cannot evolve, iterate or be subverted. This work, forthcoming at @FAccTConference ’22, critiques the existing paradigm of identity in AI. By @t1ngyu3, @jackayline & @empiricallykev. https://t.co/33LQLX0aTd
2
30
130
Really grateful to work on such important, meaningful issues. Thank you to the community of teammates, collaborators, and supporters who make this possible!
Meet Kevin, a research scientist, avid surfer, and co-founder of QueerMinds working to advocate for the LGBTQ+ community in AI research. Find out more about Kevin’s story in today’s blog: https://t.co/PuealR25d5
#LifeAtDeepMind
0
0
25
accepted into FAccT!! 👼👼👼 "subverting machines, fluctuating identities: re-learning human categorization" with @jackayline & @empiricallykev <3 so excited to share this with u all and see you in seoul 🐯 #facct22 @FAccTConference
2
4
19
Making your schedule up for #SPSP2022 tomorrow? Be sure to come by for our symposium on AI and social bias @ahbailey04 @david_m_amodio @empiricallykev 🤓🥳
0
5
9
Finally combined my passion for human-AI cooperation and my obsession with Overcooked 🧑🍳 Excited to present this project at NeurIPS with my awesome friends and teammates @djstrouse, @edwardfhughes, Matt Botvinick, and Richard Everett!
This was joint work with a set of fantastic collaborators, including my co-lead Richard Everett, @empiricallykev who led the human evaluation work, @edwardfhughes, and Matt Botvinick. Interested? Come say hi at our poster this Thu, or check out our talk. https://t.co/DkJ2cBvC2c
0
1
9
Deep RL has had lots of success producing superhuman *opponents* in games, but how do we train great *teammates*? Check out our #NeurIPS2021 Spotlight “Collaborating with Humans without Human Data”! Summary below… https://t.co/5SQ2XfL6zX
arxiv.org
Collaborating with humans requires rapidly adapting to their individual strengths, weaknesses, and preferences. Unfortunately, most standard multi-agent reinforcement learning techniques, such as...
2
20
100
"Given the historical oppression and contemporary challenges faced by queer communities, there is a substantial risk that artificial intelligence systems will be designed and deployed unfairly for queer individuals."
DeepMind researchers say AI poses a threat to people who identify as queer https://t.co/rVPgoHmyCh by @kharijohnson
0
19
33
What sort of social and/or technical efforts could help extend algorithmic fairness to LGBT+ folk?
0
0
2
Really excited to share this piece. AI creates both new opportunities and serious risks for queer communities
In a new paper, @weballergy, @jackayline, @empiricallykev & @shakir_za highlight the importance of developing new approaches for algorithmic fairness, exploring queer concerns like privacy and language to understand the effects of AI on queer communities:
1
1
15