Freddie Bickford Smith
@fbickfordsmith
Followers
447
Following
798
Media
7
Statuses
205
ML PhD at Oxford with @tom_rainforth @adamefoster. Previously at UCL with @profdata @bdroads @egrefen.
Joined June 2011
There’s a lot of confusion around uncertainty in machine learning. We argue the "aleatoric vs epistemic" view has contributed to this and present a rigorous alternative. #ICML2025 with @janundnik @eleanortrollope @markvanderwilk @adamefoster @tom_rainforth. 1/5
1
16
59
RT @willmacaskill: Today I’m releasing an essay series called Better Futures. It’s been something like eight years in the making, so I’m p….
0
40
0
Join RainML! Lots of exciting work going on :).
I have an opening for a 2-year postdoc in probabilistic machine learning and/or experimental design. The application deadline is the 3rd of September. See here for details and how to apply:
0
0
3
Strongly recommend working with Rob!.
I'm looking for talented and ambitious PhD students to join me at Nanyang Technological University Singapore to work on safe and robust AI systems!. Full scholarships covering tuition and a stipend are available, and are open to local and international students alike.
0
0
1
Come and chat at the poster session today at 4:30-7pm :). Poster E-1403, East Exhibition Hall A-B.
There’s a lot of confusion around uncertainty in machine learning. We argue the "aleatoric vs epistemic" view has contributed to this and present a rigorous alternative. #ICML2025 with @janundnik @eleanortrollope @markvanderwilk @adamefoster @tom_rainforth. 1/5
0
0
4
RT @summerfieldlab: In a new paper, we examine recent claims that AI systems have been observed ‘scheming’, or making strategic attempts to….
0
25
0
Read the paper at @adam_golinski @blackhc @davidduvenaud @rob_cornish @seb_far @sirbayes @yeewhye this is now out :). 5/5.
arxiv.org
The ideas of aleatoric and epistemic uncertainty are widely used to reason about the probabilistic predictions of machine-learning models. We identify incoherence in existing discussions of these...
0
0
13
Our view:. 1. Decision-grounded uncertainty—underappreciated in ML despite great work by @willieneis @yulantao1996 @shengjia_zhao @chenlin_meng @stefanoermon. 2. Reason about new data to define a decomposition. 3. Disambiguate uncertainty vs performance vs data dispersion. 4/5
1
0
3
This conception of uncertainty has been around for a long time—isn't it tried and tested?. In ML there are many things we want to talk about. But the aleatoric-epistemic view only has capacity for two ideas!. Using it has inevitably led to overloading and conflation. 3/5
1
1
4
It's very common to talk about two forms of uncertainty. Aleatoric ("relating to chance") is associated with statistical dispersion ("noise") in data. Epistemic ("relating to knowledge") is associated with the internal information state of a model. 2/5
1
0
5
RT @AdamEFoster: I am very happy to share Orbformer, a foundation model for wavefunctions using deep QMC that offers a route to tackle stro….
arxiv.org
Reliable description of bond breaking remains a major challenge for quantum chemistry due to the multireferential character of the electronic structure in dissociating species. Multireferential...
0
30
0
RT @MrinankSharma: Today is a big day for AI Safety. We released Claude Opus 4 under the ASL-3 deployment standard. Here's what that means:.
0
16
0
RT @stats_UCL: This week's seminar will be given by Desi Ivanova (University of Oxford). Time and date: Wednesday 26th March 2-3pm (GMT)….
0
3
0
RT @hlntnr: Lately it sometimes feels like there are only 2 AI futures on the table—insanely fast progress or total stagnation. Talked wit….
0
20
0
RT @rob_cornish: A meta-point of this paper is that category theory has utility for reasoning about current problems of interest in mainstr….
0
7
0
RT @desirivanova: 📣 Jobs alert: UQ in LLMs!. We're looking to hire a Postdoctoral Fellow and a Research Engineer to work on uncertainty qua….
0
10
0
RT @Zergylord: Illustrations like this illustrate why Jurgen hasn't been as influential as he'd like:. His ideas and papers are overly comp….
0
4
0
RT @DavidDuvenaud: New paper: What happens once AIs make humans obsolete?. Even without AIs seeking power, we argue that competitive pressu….
0
259
0
RT @seb_far: In the final stages of assembling your ICML submission? For an excellent paper, each section has a purpose and each paragraph….
0
2
0